QUANTITATIVE ANALYSIS OF PHOTOSYNTHATE UNLOADING IN

DEVELOPING SEEDS OF Phaseolus vulgaris L.

A Dissertation

Presented to the Faculty of the Graduate School

of Cornell University

in Partial Fulfillment of the Requirements for the Degree of

Doctor of Philosophy

by

Erie Christopher Ellis

January 1990

© Erie Christopher Ellis 1990

ALL RIGHTS RESERVED

QUANTITATIVE ANALYSIS OF PHOTOSYNTHATE UNLOADING IN

DEVELOPING SEEDS OF Phaseolus vulgaris l.

Erie Christopher Ellis, Ph.D.

Cornell University 1990

The pathway and kinetics of photosynthate unloading in developing

seeds of Phaseolus vulgaris were investigated using steady state labelling with

14C02' The continuous assimilation of 14C02 at a constant specific activity

was found to produce relatively stable tracer fluxes that facilitated

straightforward analyses of photosynthate import and unloading in developing

seeds. The import and partitioning of tracer within seeds was disrupted by the

surgical excision of the distal halves of seeds, as practiced during the

preparation of "empty" seed coats for perfusion. Although the perfusion of

empty seed coats produced an inhibition of photosynthate import to, and efflux

from, perfused seed coats relative to intact seeds, phloem import and

unloading of photosynthates was quantitatively significant in perfused seed

coats.

The kinetics of tracer equilibration within intact and perfused seed coats

provided evidence for the hypothesis that photosynthates imported by the

phloem are unloaded symplastically; photosynthates move throughout the

seed coat before efflux to the apoplast. Evidence for the stimulation of phloem

import by lowered cell turgor in perfused seed coats was provided by the

increase in tracer and sucrose import to seed coats treated with high

osmoticum concentrations in the perfusion solution. The partitioning of

photosynthate between retention in the seed coat and release to a perfusion

solution was turgor-sensitive, and high cell turgors were associated with an

increase in the proportion of photosynthates released to the perfusion solution

at the expense of photosynthate retention within the seed coat. Sucrose uptake

experiments demonstrated that the turgor-sensitive apoplastic retrieval

mechanism proposed by Wolswinkel and Ammerlaan (1986) was not active in

perfused seed coats. The turgor-sensitive partitioning of photosynthates

between retention in the seed coat and unloading to the apoplast was

consistent with the turgor homeostat model of photosynthate unloading as

described by Patrick et al. (1986). The efflux of unlabeled sugar and 14C_

photosynthate was stimulated by rapid changes in the osmoticum

concentration of the perfusion solution, and EDTA also stimulated

photosynthate efflux from seed coats, possibly by the direct stimulation of

photosynthate release from the phloem.

BIOGRAPHICAL SKETCH

Erie Christopher Ellis (1963- ) was born and raised in Washington,

DC, where he attended the Sidwell Friends School. Before graduating in the

spring of 1981, he experimented with hydroponics and root-less plants under

the supervision of Charles Biggs, and these experiments ultimately led to

placement in the honors group of the Westinghouse Science Talent Search.

During the summer of 1981, and for the two summers following, he worked for

Dr. Thomas E. Devine at the Cell Culture and Nitrogen Fixation Laboratory of

the USDA, ARS at Beltsville, Md. In the fall of 1981, he enrolled as an

undergraduate in the College of Arts and Sciences at Cornell University. In

1983, he began an independent study of photosynthate unloading in soybean

seed coats under the direction of Roger Spanswick, which led to an honors

thesis and to a publication (Ellis and Spanswick, 1987). After graduating Cum

Laude in Biology (AB) in January, 1986, he registered in the graduate school of

Cornell University in the fall of 1986, and began the research presented in this

dissertation in the laboratories of Drs. Roger Spanswick and Robert Turgeon.

During his three and one half years of graduate study, he was supported by

teaching assitantships, and spent two semesters teaching Introductory Botany

under Professor Karl Niklas, Plant Anatomy under Professor Dominick Paolillo,

and Plant Mineral Nutrition under Professor Leon Kochian.

iii

DEDICATION

To Tom McAllister, who taught me to respect agriculture.

The hard way.

iv

ACKNOWLEDGEMENTS

Roger Spanswick must be acknowledged for his patient dedication to

my continuing study at Cornell; without his help and encouragement I would

not have started or finished my graduate research. Robert Turgeon must also

be recognized for his material support and scientific guidance of my research.

My respect for Roger Spanswick and Robert Turgeon as human beings and as

scientists helped me to survive the rigors of graduate study at Cornell. I would

like to thank Timothy Setter for technical guidance, Richard Zobel for helping

me change my minor concentrations and Mark Sorrells for stimulating my

interest in plant breeding. I would also like to thank all three for serving on my

committee. My teaching assistanships were an educational experience thanks

to the efforts of Professors Karl Niklas, Dominick Paolillo, and Leon Kochian.

Dr. Charles McCulloch of the statistical consulting service of the Department of

Plant Breeding and Biometry assisted me with the statistics used in this

dissertation. Skilled technical assistance was provided by Esther Gowan and

the workers of the physics and chemistry machine shops at Cornell. Barbara

Bernstein produced several of the fjgures in this dissertation, and the

professors, graduate students, and secretaries of the Section of Plant Biology

helped me in countless ways. Finally, my many friends, past and present, and

particularly those I lived with collectively during most of my years as a graduate

student must be acknowledged for their endurance of my strange dedication to

legume seeds.

v

TABLE OF CONTENTS

List of Tables ................................................................................................. viii

List of Figures. ..................................... ................. ......... ................................. ix

CHAPTER 1: Introduction and Literature Review............................................. 1

Introduction ... ..................................... ................ ........................................ 1

Agronomic Importance of Photosynthate Partitioning ................ 1

General Hypotheses and Methods Used for the Study of

Photosynthate Partitioning ........................................ 4

Conclusions ............................................................................... 9

Physiological Characteristics of Photosynthate Partitioning ...................... 10

Properties of Source Leaves .................................................... 10

Properties of the Phloem Transport Path ................................. 14

Properties of Sink Organs and Tissues .................................... 16

Conclusions............. .................... ............................................ 22

Photosynthate Partitioning During the Reproductive Growth of Legumes . 24

Genetic Associations with Seed Yield ...................................... 24

Patterns of Whole Plant Photosynthate Partitioning ................. 26

Conclusions ............................................................................. 29

Photosynthate Partitioning Within Developing Legume Seeds .................. 30

The Growth and Development of Legume Seeds .................... 30

Seed Coat Structure and the Pathway of Phloem Unloading ... 32

The Concentration of Solutes in the Apoplast .......................... 34

Water Relations of the Seed Coat and Cotyledons .................. 34

Sucrose Uptake by the Developing Embryo ............................ 36

The Empty Seed Coat Technique ............................................ 38

Conclusions .............................................................................................. 49

CHAPTER 2: The Use of Steady State Labelling to Study Phloem Transport into

Developing Seeds .............. ................................................................ 51

INTRODUCTION ....................................................................................... 51

MATERIALS AND METHODS ................................................................... 53

vi

RESULTS AND DISCUSSION ................................................................... 62

CONCLUSIONS ........................................................................................ 89

CHAPTER 3: Turgor-Sensitive Photosynthate Unloading from Perfused Seed

Coats .................................................................................................. 90

INTRODUCTION ....................................................................................... 90

MATERIALS AND METHODS ................................................................... 93

RESULTS .................................................................................................. 99

DISCUSSION ...................................................................... ........ ......... ... 116

CONCLUSIONS .................................................................. ....... ....... ...... 129

CHAPTER 4: Rapid Changes in Photosynthate Unloading Induced by

Osmoticum and EDTA ...................................................................... 131

INTRODUCTION ..................................................................................... 131

MATERIALS AND METHODS ................................................................. 135

RESULTS ................................................................................................ 137

DISCUSSION ...................... .... ................................ ......... ...... ................. 150

CONCLUSIONS ................. ......................... ........... ............... ......... ......... 163

CHAPTER 5: Concluding Remarks and Directions for Future Research ..... 165

Improvement of Techniques ................................................................... 165

Directions for Future Research ............................................................... 168

Concluding Remarks ........................... ................................................... 171

LITERATURE CITED ................................................................................... 175

vii

LIST OF TABLES

2.1 - Statistical Significance of Differences Between Intact and Cut Seeds

With Respect to 14C and Sugar (sucrose + glucose)

Page

Accumulation ............................ ............................................... 83

2.2 - The Effect of Cutting on 14C and Sugar (sucrose + glucose)

Accumulation in seeds ............................................................. 85

3.1 - Effects of Osmotic Environment on Sugar (sucrose + glucose)

Elution From Seed Coats During 12 Hours of Perfusion ........ 105

3.2 - Effects of Osmotic Environment on 14C Elution From Seed Coats

During 12 Hours of Perfusion ................................................. 110

3.3 - Effects of Osmotic Environment on the Specific Activity of Sugar

(sucrose + glucose) Eluted From and Remaining in

Perfused Seed Coats ............. ................................................ 115

4.1 - The Cumulative Effects of Mannitol and EDTA Treatments on the

Quantity of Sugar (sucrose + glucose) and 14C Eluted

From and Remaining in Perfused Seed Coats ....................... 149

viii

LIST OF FIGURES

Page

2.1 - Diagram of Steady State Labelling Apparatus ....................................... 54

2.2 - Net Carbon Exchange Rate and 14C Accumulation by a Single Leaf .... 63

2.3 - 14C Partitioning in Source Leaves Over Time ....................................... 66

2.4 - Tissue Sugar (sucrose + glucose) Specific Activity Versus Time .......... 70

2.5 - Partitioning of 14C in Intact and Cut Seeds Over Time .......................... 76

3.1 - Sugar (sucrose + glucose) Efflux From Perfused Seed Coats ............ 100

3.2 - The Effects of Osmotic Environment on the Elution of Unlabeled Sugar

and 14C From Perfused Seed Coats ......................................... 103

3.3 - 14C Efflux From Attached, Perfused Seed Coats ................................ 108

3.4 - Specific Activity of Sugar (sucrose + glucose) in Perfusate From

Attached Seed Coats ................................................................. 112

3.5 - Sucrose Uptake by Detached Seed Coats ...................... ........ ........... 117

4.1 - Time Course of Seed Coat Perfusion With Changing Mannitol

Concentrations With or Without 15 mM EDTA ............................ 139

ix

4.2 - Time Course of Seed Coat Perfusion with Changing Mannitol

Concentrations With or Without 15 mM EDTA ............................ 141

4.3 - Time Course of Seed Coat Perfusion With Changing Mannitol

Concentrations With or Without 15 mM EDTA ............................ 143

4.4 - Time Course of Seed Coat Perfusion With Changing Mannitol

Concentrations With or Without 15 mM EDTA ............................ 145

x

CHAPTER 1

Introduction and Literature Review

INTRODUCTION

The growth, development, and reproduction of higher plants requires

precise regulation of the metabolism, storage, and processing of the products

of photosynthetic carbon fixation. Much of this regulation is accomplished

through the chemical and/or spatial partitioning of these compounds within the

substrate pools, organelles, cells, tissues, and organs of higher plants.

Mechanisms regulating the partitioning of photosynthates are complex, and

potential control points may be present at every level of biological complexity,

from individual genes and enzymes to whole plant systems. The importance of

photosynthate partitioning between plant parts to the agronomic yield of crop

plants has been well established, although the mechanisms regulating this

partitioning are not well understood. This chapter presents a review of current

photosynthate partitioning literature relating to the physiological mechanisms

which interact to produce yield in grain legumes. The mechanisms controlling

photosynthate partitioning in developing legume seeds will be emphasized,

and gaps in our understanding of these mechanisms will be defined so that

further investigations may be designed. In order to simplify discussion of the

many different properties of photosynthate partitioning, this review will focus on

the most widely accepted hypotheses in the literature.

Agronomic Importance of Photosynthate Partitioning

The yield of crop plants has been increased both by modifications in the

crop environment, and by improvements in genetic potential. Heritable

improvements in yield potential have resulted from crop adaptation to

enhanced growing conditions and from increases in the biological efficiency of

1

2

crop processes which are responsible for the production of agronomically

useful plant parts. The relative importance of genetically improved biological

efficiency in the development of modern high-yielding cultivars has been

difficult to assess, but there is diverse evidence supporting the view that

improvements in biological efficiency have been responsible for significant

increases in crop yield (Gifford and Evans, 1981; Gifford et aI., 1984; Snyder

and Carlson, 1984; Nelson, 1988). Attempts to increase crop yield by selection

for high photosynthetic rates on a leaf area basis have generally failed (Gifford

and Evans, 1981; Gifford et aI., 1984; Nelson, 1988), although there is some

evidence that selection for whole-canopy photosynthesis during the period of

reproductive growth may be more effective (Wells et aI., 1982; Ashley and

Boerma, 1989). The highest yielding cultivars often have low photosynthetic

rates on a leaf area basis (Peet et al. 1977; Nelson, 1988), which leads to the

conclusion that high yield must be the result of selection for plant processes

other than photosynthetic rate (Nelson, 1988). In fact, there is considerable

evidence which indicates that the most characteristic feature of modern high­

yielding cultivars is their high harvest index (HI), or ratio of economic product to

above ground biomass (Snyder and Carlson, 1984; Gifford, 1986; Nelson,

1988; Gent et aI., 1989). A high harvest index may contribute to the harvestable

yield of crop plants by increasing the efficiency of mechanical harvesting.

However, the efficency of photosynthate partitioning between harvestable yield

and other biomass may also be related to the biological efficency of yield

production in crop plants.

The efficiency with which a given cultivar partitions its biomass between

economically useful products and other biomass will depend on both

environmental and genetic factors, although it appears that harvest index is a

relatively stable indicator of cultivar performance under diverse environmental

3

conditions, at least in soybean (Snyder and Carlson, 1984; Spaeth et aI., 1984).

Breeding programs focused on selection for improved harvest index mayor

may not produce improvements in harvestable yield for several reasons, the

most important of which is that selection for a single characteristic typically

leads to compensatory changes in other important characteristics (Snyder and

Carlson, 1984; Rasmusson, 1987; Nelson, 1988). Some of the difficulties in

producing higher yielding cultivars by breeding for high harvest index may be

surmounted by simultaneous selection for both high harvest index and high

biological yield. However, it is often observed that selection for harvest index

correlates with modifications in other important plant characteristics such as

growth duration and yield quality (Snyder and Carlson, 1984; Rasmusson,

1987; Nelson, 1988).

The mechanisms governing partitioning are "plastic", in that they often

respond to modifications in the efficiency of specific processes by

compensatory changes in other processes (Gifford et aI., 1984; Snyder and

Carlson, 1984; Geiger, 1987; Nelson, 1988). The difficulties encountered in

attempts to improve yield by the selective modification of general traits such as

harvest index or photosynthetic rate points out the need for an integrated,

multi-disciplinary approach to the improvement of photosynthate partitioning in

crop plants (Gifford and Evans, 1981; Gifford et aI., 1984). Without knowledge

of the interactive properties of the different mechanisms that are responsible for

regulating photosynthate partitioning, it is unlikely that selective modification of

individual photosynthate partitioning mechanisms will lead to higher yielding

crops (Geiger, 1986, 1987). Due to the complexity of whole plant systems in

regulating the patterns of photosynthate partitioning, portions of whole plant

systems must be studied separately and the results integrated (Kursanov,

4

1984; Gifford et aI., 1984; Geiger, 1986, 1987). The purpose of this chapter is to

provide a background for the evaluation of the physiological mechanisms

which regulate photosynthate partitioning in developing legume seeds so that

the importance of this portion of the whole plant partitioning system may be

understood, and investigated.

General Hypotheses and Methods Used for the Study of

Photosynthate Partitioning

The source-path-sink model. The most basic hypothesis of whole­

plant photosynthate partitioning is the source-path-sink model (Warren-Wilson,

1972). In this system, there is a source of photosynthate (typically a mature

leaf) connected via the phloem "path" with a sink for photosynthate, which may

be almost any tissue in the plant, depending on developmental stage and

environmental conditions. The most important characteristic of a source is its

ability to produce a net export of organic materials, and thus many organs

including roots and stems may at some point be classified as sources. Sinks

are net importers of organic compounds, and the phloem transport path is a

region through which materials pass without major losses due to metabolism

and storage along the way. With the exception of leafy vegetables, plant parts

harvested for human consumption (seeds, fruits, roots, and tubers) are

typically classified as sinks (Ho, 1988).

In the simplest form of the source-path-sink model, partitioning between

tissues and organs is caused by differentials in source and sink "strength", with

the phloem path serving only as an open conduit connecting source and sink

(Warren-Wilson, 1972). The partitioning of photosynthate among various sinks

is thus the result of competition for photosynthates produced by the source

due to differences in the strength of individual sinks (Warren-Wilson, 1972; Ho,

5

1988). The concepts of source and sink strength are based on mass balance

equations which describe the net accumulation of dry weight in sinks and the

net export of materials from sources. The use of experimental manipulations

which increase the relative capacity of the source to export photosynthate to a

particular sink (C02 enrichment, light enhancement, removal of competing

sinks) has led to the concepts of source versus sink limitation (reviews by

Patrick, 1988; Ho, 1988). When the capacity for export from the source is

increased, with no observable increase in photosynthate accumulation by the

sink, then the yield of the plant part under study is said to be sink-limited. If

photosynthate accumulation in the sink is increased by enhanced export

capacity in the source, then source limitation is presumed. The relative sink

strength of various competing sinks may also be manipulated and studied by

similar methods (Ho, 1988).

The concepts of source and sink strength and limitation are useful for

describing the patterns of assimilate distribution between plant parts, and may

aid in the observation of changes in the patterns and properties of assimilate

distribution when comparing different environmental or genetic conditions.

Developmental patterns of partitioning are responsible for the duration and

timing of events such as leaf expansion, flowering, seed set, and seed fill, and

may also influence the rates at which these events proceed (Gifford and Evans,

1981). Although developmental events and genetic and environmental

constraints may be crucial in determining the patterns of photosynthate

partitioning between different parts of the plant, the physiological mechanisms

that produce these patterns cannot be understood solely through the analysis

of assimilate partitioning between different plant parts (Gifford and Evans.

1981; Lang and Thorpe, 1983; Geiger, 1987). This review will focus on the

6

physiological regulation of the rates of photosynthate partitioning mechanisms,

and will discuss the duration and timing of alterations in photosynthate

partitioning patterns only where pertinent to the description of mechanisms

which regulate partitioning rates.

Physiological components of partitioning systems. The

mechanisms which regulate the movement and transformation of

photosynthates are constrained by the anatomy, physiology, and biochemistry

of the cells, tissues, and organs that together form the whole plant. The

concepts of sink and source strength are inadequate for the full description of

these partitioning mechanisms, and more complex analyses that address

relatively specific levels of photosynthate partitioning mechanisms have been

developed by many researchers (Fisher, 1970c; Keener et aI., 1979; Herold,

1980; Gifford and Evans, 1981; Geiger et aI., 1983; Lang and Thorpe, 1983;

Shawet aI., 1986; Geiger, 1987; Huber and Kerr, 1987; Wann and Raper, 1987;

Patrick, 1988; Ho, 1988; Minchin and Grusak, 1988). The pool:process

concept of photosynthate partitioning developed by Lang and Thorpe (1983) is

oased on a system of equations that were theoretically derived from

biochemical and physiological mechanisms. Analyses based directly on

measurements of substrate compartmentation, flux, and biochemical

transformation have been independently developed by many workers (Fisher,

1970c; Keener et aI., 1979; Herold, 1980; Geiger et aI., 1983; Shaw et aI., 1986;

Geiger, 1987; Huber and Kerr, 1987; Wann and Raper, 1987; Patrick, 1988).

The anatomy and ultrastructure of many organs and tissues have been

studied in terms of their ability to facilitate symplastic and/or apoplastic

movement of photosynthates between the cells and tissues which produce,

utilize, store, and transport photosynthate (Felker and Shannon, 1980; Gifford

7

and Evans, 1981; Ungle and Chevalier, 1984; Offler and Patrick, 1984;

Schmalstig and Geiger, 1985; Thorne, 1985; Offler and Patrick, 1986;

Schmalstig et aI., 1986; Wolswinkel, 1987b; Oparka and Prior, 1988; Ho, 1988).

Anatomical studies are very useful for delimiting possible pathways of

photosynthate movement, but are not capable of conclusive differentiation

between various mechanisms and pathways of photosynthate partitioning.

Analyses of the biochemical processes which facilitate the production,

transformation, and storage of photosynthates have attempted to describe the

metabolic regulation of photosynthate partitioning by the activities of enzymes

and the fluxes of various substrates between different metabolic pathways

(Keener et aI., 1979; Herold, 1980; Ho, 1986; Huber et aI., 1986; Stitt, 1986;

Huber and Kerr, 1987; Plaut et aI., 1987; Foyer, 1988; Rocher, 1988; Servaites

et aI., 1989; Sung et aI., 1989). Photosynthate fluxes may also be regulated by

the biophysical properties of the pathways of photosynthate movement such

as substrate concentration and turgor pressure differentials between various

cells and tissues (Fisher, 1970c; Fisher, 1978; Lang and Thorpe, 1983; Fondy

and Geiger, 1983; Aloni et aI., 1986; Fisher and Gifford, 1986; Lang et aI., 1986;

Lang and Thorpe, 1986; Murphy, 1986; Wyse et aI., 1986; Daie, 1987a; Minchin

and Thorpe, 1987a; Porter et aI., 1987a; Oparka and Wright, 1988; Patrick,

1988; Estruch et aI., 1989).

The search for limiting factors in photosynthate partitioning mechanisms

has led to the "bottleneck" approach to partitioning analysis (Gifford et aI.,

1984; Kursanov, 1984; Ho, 1988). In this approach, individual steps in

partitioning systems are evaluated in order to determine whether they are the

limiting step to the overall partitioning process. In practice this approach is

limited by the complexity and plasticity of the mechanisms regulating

8

photosynthate partitioning because the enhancement of the "limiting step" in a

partitioning process often leads to another step becoming a limiting factor

(Gifford et aI., 1984; Geiger, 1987). Thus, a realistic approach to the

enhancement of photosynthate partitioning involves an understanding that no

single factor may be the limiting factor, and that real improvements in

partitioning will only result from a knowledge of how various steps in a pathway

interact to regulate partitioning (Geiger, 1987; Rocher, 1988).

The Use of Tracers. Detailed investigations of the pools and

processes involved in photosynthate partitioning have been made possible by

the use of radiotracers to monitor the fluxes and transformations of carbon­

containing compounds in vitro and in vivo. In vitro techniques for the study of

partitioning mechanisms facilitate straightforward methods for tracer

introduction to the system under investigation. The introduction of labeled

carbon into plants in vivo is more complex, and analyses of tracer movement

throughout the plant are often limited by the tracer methods that are employed

(Geiger, 1980; Zierler, 1981; Kouchi and Yoneyama, 1984a; Geiger and Shieh,

1988). Tracer carbon may be introduced to intact plants by either short term

pulse-labelling of leaves with labeled C02, or by the long term assimilation of

labeled C02 by steady state labelling methods. Pulse-labelling of leaves with

labeled C02 provides a simple, rapid method for tracer introduction, however,

this method is limited by the complexity and dynamics of short "pulses" of

tracer as they move through more and more substrate pools following

assimilation (Zierler, 1981; Geiger and Shieh, 1988). Steady state labelling of

leaves with labeled C02 is a more complex and time-consuming method.

However, tracer kinetics are simplified due to the production of relatively stable

tracer fluxes from pool to pool in the plant (Geiger, 1980; Kouchi and

9

Yoneyama, 1984a; Geiger and Shieh, 1988). Pulse-labelling studies are useful

for qualitative analyses of photosynthate partitioning patterns in whole plants

(Sakri and Shannon, 1975; Lucas et aI., 1976; Wien et aI., 1976; Heitholt et aI.,

1986; Jaeger et aI., 1988), or kinetic and compartmental analyses of

photosynthate partitioning in limited portions of whole plant partitioning

systems (Fisher, 1970a, 1970b; Hoddinott and Jolliffe, 1988; Minchin et aI.,

1984; Ntsika and Delrot, 1986; Hayes et aI., 1987; Minchin and Thorpe, 1987a,

1987b; Plaut et aI., 1987; Rocher, 1988; Thorpe and Minchin, 1988). Steady

state labelling methods facilitate quantitative high resolution measurements of

photosynthate movement between source and sink organs (Kouchi and

Yoneyama, 1984a; Gordon, 1986; Marowitch et aI., 1986; Yamagata et aI.,

1987; Geiger and Shieh, 1988) and between substrate pools (Fondy and

Geiger, 1982; Geiger et aI., 1983; Kouchi and Yoneyama, 1984b; Grange,

1985; Geiger et aI., 1983; Gordon, 1986; Shawet aI., 1986; Fondy et aI., 1989).

Conclusions

The inherent complexity and plasticity of partitioning mechanisms make

it difficult to determine the significance of individual limiting steps to the overall

limitation of photosynthate partitioning. Therefore, the purposeful improvement

of photosynthate partitioning must take into account the integration of many

individual processes which together produce the net regulation of

photosynthate partitioning. Pulse-labelling studies are limited with respect to

quantitative analyses of whole plant partitioning systems. Steady state labelling

methods allow quantitative analyses of whole plant photosynthate fluxes at the

level of tissues and organs, or at the level of individual substrate pools, and

may therefore be essential for integrated analyses of whole plant

photosynthate partitioning mechanisms.

10

PHYSIOLOGICAL CHARACTERISTICS OF CARBON PARTITIONING

Properties of Source Leaves

The most important property of source leaves in relation to whole plant

photosynthate partitioning is their ability to export photosynthates to sinks. The

net production of photosynthate occurs during the light portion of the

photoperiod in most crop plants, even though substantial photosynthate export

takes place 24 hours a day (Fondy and Geiger, 1982, 1983; Gordon, 1986;

Shaw et aI., 1986; Mullen and Koller, 1988a, 1988b). For this reason,

photosynthates must be stored for export during the dark period when

photosynthates are not produced (Fondy and Geiger, 1982, 1983; Gordon,

1986; Shaw et aI., 1986; Mullen and Koller, 1988a, 1988b). Photosynthate

allocation between export and storage depends on the regulation of substrate

compartmentation within the leaf, and different plants have different

mechanisms for controlling this process (Gordon, 1986, Huber et aI., 1986;

Huber and Kerr, 1987). The most common photosynthate exported from the

source leaves of crop plants (excluding the Cucurbitaceae and several other

families) is sucrose, and either starch, or both sucrose and starch, may be

stored for future mobilization and export (fructans may also be stored in the

gramineae) (Gordon, 1986; Huber et aI., 1986; Huber and Kerr, 1987).

Photosynthetic carbon fixation occurs in the stroma of chloroplasts, and

carbon flux between the chloroplasts and the cytosol is probably mediated by

the phosphate trans locator which exchanges triose phosphates for inorganic

phosphate (Herold, 1980; Huber et aI., 1986; Stitt, 1986; Huber and Kerr,

1987). During the light period, triose phosphates are released from the

chloroplasts to the cytoplasm where sucrose is synthesized by a pathway with

11

a complex regulatory system that has been shown to be controlled primarily by

the coordinated metabolic regulation of two key enzymes: sucrose-phosphate

synthase (SPS), and Fru-1,6-BPase (Huber et aI., 1986; Huber and Kerr, 1987).

The rate of sucrose synthesis in the cytosol for export during the light period

appears to be regulated by the activity of SPS in dicots, but may be regulated

by more complex mechanisms in monocots (Huber et aI., 1986). Sucrose

synthesis during the dark period does not appear to be regulated by SPS

activity, and may be limited by starch breakdown (Huber and Kerr, 1987),

although this process is not fully understood (Del rot and Bonnemain, 1985;

Grange, 1985; Fondy et aI., 1989; Servaites et aI., 1989).

Because starch is stored in the chloroplasts, mobilization for export at

night (or during peak demand by sinks) most likely occurs by the breakdown of

starch into triose phosphates which are transported to the cytosol where

sucrose is synthesized (Delrot and Bonnemain, 1985; Huber et aI., 1986;

Huber and Kerr, 1987; Foyer, 1988). However, recent studies by Servaites et

al. (1989) and Fondy et al. (1989) suggest that an alternative pathway for

starch breakdown and sucrose synthesis may also exist. In the gramineae and

other plants which store both sucrose and starch for later export (which may

also occur to a limited degree in many other plants), sucrose is sequestered in

the vacuole during the light period and then transported to the cytoplasm for

export (Geiger et aI., 1983; Delrot and Bonnemain, 1985; Gordon, 1986; Foyer,

1988).

Sucrose mobilization for export may be regulated by endogenous

rhythms in enzyme activity (Fondy and Geiger, 1982, Huber et aI., 1986; Huber

and Kerr, 1987). In addition, sucrose mobilization may respond to changes in

the balance between photosynthate production and sink demand for

12

photosynthates (Herold, 1980; Azcon-Bieto, 1983; Carlson and Brun, 1985;

Huber and Kerr, 1987; Plaut et aI., 1987; Mullen and Koller, 1988a; Foyer,

1988). The modification of photosynthate export caused by changes in sink

demand may be regulated by end product inhibition of photosynthesis caused

by changes in sink demand (rhorne and Koller, 1974; Clough et aI., 1981;

Azcon-Bieto, 1983; Foyer, 1988). Sink demand may also alter the allocation of

fixed carbon between sucrose production for export and the storage of starch

in the leaf, with or without the necessity of changes in the photosynthetic rate,

or even in the rate of export (rhorne and Koller, 1974; Fondy and Geiger, 1980;

Herold, 1980; Carlson and Brun, 1985; Grange, 1985; Plaut et aI., 1987).

Carbon allocation between starch storage versus sucrose synthesis may be

coordinated with changes in sink demand by a mechanism involving the

exchange of inorganic phosphate between the chloroplasts and the cytosol as

proposed by Herold (1980), although recent evidence has complicated this

hypothesis (Fondyet aI., 1989; Servaites et aI., 1989). Regardless of the

mechanism by which sucrose synthesis for export is regulated, there is

voluminous evidence that sink demand may influence leaf export in the short

term (Fondy and Geiger, 1980; Geiger and Fondy, 1980; Azcon-Bieto, 1983;

Kallarackal and Milburn, 1984; Carlson and Brun, 1985; Ntsika and Delrot,

1986; Plaut et aI., 1987; Foyer, 1988). There is also evidence that increased

sink demand may enhance export from the leaf in the long term (rhorne and

Koller, 1974; Pereto and Beltran, 1987).

Sucrose synthesized in the cytosol of leaf mesophyll cells must be

loaded into the phloem before export from the leaf. The pathway of

photosynthate movement from the mesophyll to the lumen of the phloem may

be either symplastic (Kaiser and Martinoia, 1985), apoplastic (review by Delrot,

13

1987), or both symplastic and apoplastic (Thorpe and Minchin, 1988). The

pathway of phloem loading may be different in different plants (review by

Delrot, 1987). There is a great deal of evidence for phloem loading from the

apoplast, while the evidence for symplastic phloem loading remains

speculative (Delrot, 1987). Regardless of whether phloem loading is apoplastic

or symplastic, high concentrations of sucrose and other osmotically active

substances are present in the phloem relative to the mesophyll in mature,

exporting source leaves (Fisher, 1978; Geiger and Fondy, 1980; Delrot, 1987).

Thus, the hypothesis of symplastic phloem loading is complicated by the lack

of a well documented mechanism by which substances can be concentrated

across cells linked by plasmodesmata (Delrot, 1987). Phloem loading requires

energy (Fondy and Geiger, 1983; Daie, 1987a, 1987b; Thorpe and Minchin,

1988), and appears to be carrier mediated under some conditions (Daie,

1987a, 1987b; Aloni et aI., 1988; Estruch et aI., 1989), but not in others (Thorpe

and Minchin, 1988). Phloem loading of sucrose from the apoplast, appears to

be facilitated by a sucrose/proton symport driven by the electrochemical

potential gradient for protons generated by a plasmalemma H + -ATPase (Daie,

1987a, 1987b; Aloni et aI., 1988; Estruch et aI., 1989).

The hypothesis that the process of phloem loading may regulate

photosynthate export from leaves has not been proven, and this mechanism

does not appear to limit export (Kallarackal and Milburn, 1984; Delrot and

Bonnemain, 1985). There is now convincing evidence that decreasing phloem

turgor may enhance phloem loading (Daie, 1987a; Aloni et aI., 1988; Estruch et

aI., 1989). Thus, sink demand may regulate phloem loading and export by

modifying phloem turgor (Lang, 1983; Kallarackal and Milburn, 1984; Lang and

Thorpe, 1986; Daie, 1987a; Minchin and Thorpe, 1987a; Patrick, 1988; Estruch

14

et aI., 1989). Phloem loading may also be influenced by growth regulators

endogenous to the leaf or by those produced by sinks and transported to the

leaf (Lenton, 1984; Daie, 1987a; Pereto and Beltran, 1987; Aloni et aI., 1988;

Patrick, 1988; Estruch et aI., 1989).

Properties of the Phloem Transport Path

The most well accepted model for photosynthate transport via the

phloem is the mass flow hypothesis, in which the driving force for phloem

transport is provided by a turgor pressure differential in the sieve tubes

between source and sink (Munch, 1930; Goeschl et aI., 1976; Fisher, 1978;

Lang, 1983; Delrot and Bonnemain, 1985; Lang et aI., 1986; Lang and Thorpe,

1986; Minchin and Thorpe, 1987a). The turgor pressure of sieve tubes at the

source is higher than in the sink end of the pathway, and thus the mass flow of

phloem sap may occur down a turgor pressure gradient from source to sink

(Fisher, 1978; Delrot and Bonnemain, 1985). The high turgor pressure of

source sieve tubes is probably generated by the active loading of osmotically

active substances into the lumen of the phloem (Gifford and Evans, 1981; Daie,

1987a, 1987b; Aloni et aI., 1988; Estruch et aI., 1989), which causes the influx

of water and the development of high turgor (Fisher, 1978; Geiger and Fondy,

1980; Delrot and Bonnemain, 1985; Delrot, 1987). At the sink end of the

pathway, materials are unloaded from the phloem, thus producing a lower

turgor pressure at the sink (Fisher, 1978; Wolswinkel, 1985b; Lang et aI., 1986;

Lang and Thorpe, 1986). The process of phloem unloading in sinks will be

discussed in the next section of this review.

The most highly concentrated substances in the phloem are sugars

(primarily sucrose in most crop plants) and potassium (Fisher, 1970a; Gifford

and Evans, 1981; Lang, 1983). A model which depends on the coordinated

loading and unloading of sugar and potassium for the regulation of mass flow

15

has been developed by Lang (1983). In this variation on the mass flow

hypothesis as originally proposed by Munch (1930), the regulation of

potassium gradients between the phloem and xylem causes changes in

phloem turgor that allow the enhancement or retardation of mass flow from

source to sink (Lang, 1983). This model thus allows the regulation of phloem

transport by mechanisms other than the export and uptake of sugars in the

source and sink respectively (Lang, 1983). The physiological significance of

this model has not been tested and, in general, translocation appears to be

constrained by the turgor differentials maintained by phloem loading of sugars

at the source and phloem unloading at the sink (Lang, 1983; Kursanov, 1984;

Lang and Thorpe, 1986; Lang et aI., 1986).

Long distance phloem transport of photosynthates to sinks without

major losses along the way is an important function of path tissues (petioles,

stems, and peduncles). The high concentration of solutes in the phloem would

be expected to promote the passive leakage of these materials along the path

of photosynthate transport (Minchin et aI., 1984; Aloni et aI., 1986; Wolswinkel,

1987b). In spite of the potential for passive leakage from the phloem, the

majority of photosynthates transported through path tissues remain in the

phloem under most conditions (Fisher, 1970a; Minchin et al., 1984; Delrot and

Bonnemain, 1985). There is considerable evidence that in stems, phloem

leakage to the apoplast is countered by the active reloading of sucrose and

other materials into the phloem from the apoplast (Minchin et aI., 1984; Hayes

et aI., 1987; Minchin and Thorpe, 1987b; Patrick and Mulligan, 1989). There is

also evidence that photosynthates may move symplastically from the phloem

to other cells in stems under sink-limited conditions (Hayes et aI., 1987). The

materials unloaded from the phloem in path tissues may act as a "phloem

16

buffering pool" that minimizes changes in phloem transport caused by rapid

alterations in export from the source or import by the sink (Minchin et aI., 1984;

Franceschi, 1986; Hayes et aI., 1987; Patrick and Mulligan, 1989). The phloem

buffering pool is primarily apoplastic (Minchin et aI., 1984; Hayes et aI., 1987;

Patrick and Mulligan, 1989). However, under conditions of sink limitation, the

symplastic pool may also contribute to phloem buffering (Hayes et aI., 1987).

The extent of the phloem buffering pool may also be regulated by hormonal

treatments (Patrick and Mulligan, 1989).

Properties of Sink Organs and Tissues

Before photosynthates imported to sinks may be used for growth

and/or storage, these materials must be unloaded from the phloem to other

tissues in the sink (Ho, 1988). The mechanism of phloem unloading, and the

properties of photosynthate utilization vary from sink to sink and from plant to

plant, and may also depend on developmental stage and environmental

influences (Thorne and Giaquinta, 1984; Offler and Patrick. 1986; Ho, 1988;

Patrick; 1988). For the purposes of this review, sinks will be classified as either

meristematic sinks (root and shoot apices, and developing leaves), reversible

storage sinks (storage roots and tubers), or irreversible storage sinks

(developing seeds) (after Offler and Patrick, 1986). For the sake of brevity,

fleshy fruits and stems will not be discussed. Photosynthate unloading in

developing legume seeds will be discussed at length in a separate section of

this review, due to the quantity of research that has focused on this system.

Meristematic sinks. Phloem unloading in meristematic sinks appears

to be facilitated by the symplastic movement of photosynthates from the

phloem to other cells (Thorne and Giaquinta, 1984; Delrot and Bonnemain,

1985; Schmalstig and Geiger, 1985; Wolswinkel, 1985b; Schmalstig et aI.,

17

1986; Thorne, 1986; Ho, 1988). The movement of photosynthates into

meristematic sinks is probably sustained by a concentration gradient between

the phloem and sink cells caused by the rapid utilization of materials for growth

(Delrot and Bonnemain; 1985; Schmalstig and Geiger, 1985; Ho, 1988).

Potato tubers. Ultrastructural and fluorescent dye microinjection

studies of developing potato tubers have led to the hypothesis that the

movement of photosynthates from the phloem to the cortical storage cells is

symplastic (Oparka, 1986; Oparka and Prior, 1988). Further evidence for

symplastic phloem unloading in potato tubers was provided by plasmolysis

experiments (Oparka and Prior, 1987). Wright and Oparka (1989) have

proposed that, following symplastic unloading from the phloem, the majority of

unloaded sucrose is converted to starch in the amyloplasts, with additional

storage of sucrose in the vacuole. The symplastic movement of sucrose is thus

driven by the concentration gradient between the phloem and the cortical

storage cells generated by the rapid accumulation of starch in the amyloplasts

(Oparka, 1986; Wright and Oparka, 1989). Sucrose uptake into potato tuber

disks appears to be carrier-mediated and was enhanced by decreasing cell

turgor, while starch synthesis was optimized by a 300 mM osmoticum

treatment (Oparka and Wright, 1988a, 1988b). The purpose of an active

(carrier-mediated) sucrose retrieval mechanism in the plasmalemma of storage

cells is to recover sucrose that is passively leaked to the apoplast (Wright and

Oparka, 1989). The increases in sucrose uptake and starch storage caused by

low cell turgors in potato tubers may facilitate the coordination of carbohydrate

accumulation in the tuber with the rate of phloem import to the tuber, which

may be increased by low phloem turgor (Wolswinkel, 1985b; Lang and Thorpe,

1986).

18

Sugar beet storage roots. The pathway of phloem unloading in sugar

beet storage roots has been hypothesized to consist of direct phloem

unloading of sucrose to the apoplast followed by active sucrose uptake into

storage parenchyma cells where sucrose is accumulated in the vacuole (review

by Ho, 1988). The uptake of sucrose from the apoplast appears to be mediated

by a sucrose/proton symport that is stimulated by low cell turgor rNYse et aI.,

1986). The turgor-sensitivity of the sucrose/proton cotransport mechanism

may be due to the inhibition of an electrogenic proton pump by high cell turgor

(Kinraide and Wyse, 1986). Turgor-sensitive sucrose uptake by the storage

parenchyma may facilitate the coordination of sucrose uptake with phloem

unloading to the apoplast because, at high apoplastic sucrose concentrations,

phloem import is probably stimulated (Wolswinkel, 1985b; Lang and Thorpe,

1986), as is sucrose uptake (Wyse et aI., 1986). The ultimate sink for sucrose in

the sugar beet storage root is the vacuole, where sucrose is accumulated to

very high concentrations by the action of a sucrose/proton antiport (Briskin,

1986). It has been hypothesized that the active, carrier-mediated uptake of

sucrose across the tonoplast provides the driving force for sugar accumulation

in the storage root (Briskin, 1986).

Maize kernels. Photosynthates (primarily in the form of sucrose) move

symplastically from the phloem to the placento-chalazal cells in the pedicel of

developing maize kernels (Felker and Shannon, 1980). Sucrose is then

released to the apoplast where it is hydrolyzed to hexoses (glucose and

fructose) by extracellular invertases prior to hexose uptake by the adjoining

endosperm transfer cells (Felker and Shannon, 1980; Porter et aI., 1985;

Griffith et aI., 1986; Shannon et aI., 1986). The translocation of photosynthates

into the pedicel was reduced by treatment with inhibitors, although efflux to the

19

apoplast did not appear to be sensitive to inhibitors (Porter et aI., 1985, Porter

et aI., 1987b). The results of these experiments support a passive mechanism

of sucrose efflux to the apoplast, driven by the inversion of sucrose followed by

the uptake of hexoses and the resynthesis of sucrose in the endosperm and

embryo (Griffith et aI., 1986; Shannon et aI., 1986). The uptake of sugars from

the apoplast may be mediated by a hexose/proton cotransport mechanism in

maize embryos, while hexose uptake into the endosperm appears to be

passive (Griffith et aI., 1986; Griffith et aI., 1987a, 1987b). The mechanism of

photosynthate unloading in the maize pedicel appears to be turgor-sensitive,

because higher cell turgors promote photosynthate efflux to the apoplast

(Shannon et aI., 1986; Porter et aI., 1987a). The concentration of sugars in the

apoplast of the maize pedicel parenchyma-placento-chalazal tissue is quite

high (between 470 and 800 mM) and it has been hypothesized that turgor­

sensitive photosynthate unloading from the pedicel may thus be regulated by

the concentration of sugars and other assimilates in the apoplast (Shannon et

aI., 1986; Porter et aI., 1987a). In addition, the uptake of sucrose and amino

acids by the maize embryo (and possibly by the endosperm) appears to be

stimulated by low cell turgor (Wolswinkel and Ammerlaan, 1989). Unloading

from the pedicel may thus be coordinated with uptake by the embryo and

endosperm through turgor regulation based on the concentration of sugars in

the apoplast, because unloading is inhibited while uptake is enhanced by low

cell turgor (Wolswinkel and Ammerlaan, 1989). It has been hypothesized that

the turgor-sensitive photosynthate unloading mechanism allows the rate of

assimilate transport to the kernel, which may be stimulated by low cell turgor

20

(review by Wolswinkel, 1985b; Lang and Thorpe, 1986), to be controlled by the

rate of sugar accumulation in the endosperm and embryo, through changes in

the concentration of sugars in the apoplast (Shannon et aI., 1986; Porter et al.,

1987a).

Sorghum caryopses. The pathway of phloem unloading in sorghum

appears to be similar to maize, in that photosynthates (primarily in the form of

sucrose) move symplastically from the phloem to the placental tissues where

efflux to the noncellular placental sac region occurs. (Maness and McBee,

1986). Also, sucrose is inverted to hexoses prior to uptake from the solution in

the placental sac region by endosperm transfer cells (Maness and McBee,

1986).

Wheat and barley grains. Photosynthates (primarily in the form of

sucrose) are transported throughout the entire length of the crease region in

wheat and barley grains via the phloem (Thorne 1985). Sucrose is then

symplastically unloaded to the pigment strand parenchyma followed by

symplastic movement to the transfer cells of the nucellar projection where

release to the noncellular endosperm cavity probably occurs (Sakri and

Shannon, 1975; Cook and Oparka, 1983; Thorne 1985). Sucrose is not

significantly hydrolyzed before uptake by the aleurone transfer cells adjoining

the endosperm cavity and, following uptake, sucrose moves throughout the

endosperm where starch synthesis occurs (Sakri and Shannon, 1975; Lingle

and Chevalier, 1984; Thorne, 1985). There was a significant sugar

concentration gradient across the crease region between the phloem and the

endosperm cavity (Fisher and Gifford, 1986), indicating that accumulation and

storage in the endosperm (as starch) may have been limited by sugar efflux

from the phloem (Jenner and Rathjen, 1978; Ho and Gifford, 1984; Gifford,

21

1986). Thus, carbohydrate movement to the endosperm may be limited by the

transfer of assimilates from the sieve tubes to the tissues of the crease region

(Ho and Gifford, 1984; Fisher and Gifford, 1986). However, there is evidence

for the limitation of carbohydrate accumulation in wheat grains by the rate of

sucrose conversion to starch in the endosperm (Jenner and Rathjen, 1978;

Lingle and Chevalier, 1984; Martinez-Carrasco et aI., 1988).

The driving force for sugar movement from the phloem to the

endosperm cavity appears to be the concentration gradient between the

phloem and the endosperm cavity generated by sucrose uptake and starch

formation in endosperm storage cells (Fisher and Gifford, 1986). The uptake of

sucrose for starch formation by the endosperm appears to be passive,

because the concentration of sucrose in the endosperm is lower than in the

endosperm cavity (Ho and Gifford, 1984). However, the inhibition of sucrose

uptake by PCM BS demonstrated that a sucrose carrier mechanism may be

present at the endosperm plasmalemma (Ho and Gifford, 1984). Although

evidence for sucrose uptake by a sucrose/proton cotransport mechanism was

presented by Ho and Gifford (1984), the concentration gradient between the

endosperm cavity and the endosperm indicates that sucrose uptake by the

endosperm is passive. Thus, the driving force for sugar transport into the

developing grains of wheat and barley is the downhill concentration gradient

for sugar movement generated by starch storage in the endosperm.

Rice caryopses. The pathway of photosynthate unloading in rice

caryopses is similar to that of wheat and barley in that photosynthates move

symplastically from the phloem to the nucellus (Oparka and Gates, 1984;

Thorne, 1985). In contrast with wheat, barley,· and maize, the symplastic

movement of photosynthates within the nucellus may facilitate the movement of

22

photosynthates around the entire caryopsis in rice (Oparka and Gates, 1984;

Thorne, 1985). Therefore photosynthate unloading from the nucellus to the

apoplast at the nucellus/aleurone interface may occur over the entire surface

of the aleurone and endosperm in the developing rice caryopsis (Oparka and

Gates, 1984; Thorne, 1985).

Conclusions

The production and export of photosynthates from source leaves

appears to be coordinated with photosynthate import by sinks, although the

mechanisms which bring about this regulation are complex and are not well

understood (Herold, 1980; Fondy and Geiger, 1983; Lang, 1983; Kursanov,

1984; Delrot and Bonnemain, 1985; Wolswinkel, 1985b; Geiger, 1986, 1987;

Lang and Thorpe, 1986; Pereto and Beltran, 1987). Although not discussed

explicitly above, there is abundant evidence for the regulation of photosynthate

partitioning by the hormonal control of both source and sink strength (Herold,

1980; Gifford and Evans, 1981; Herzog, 1982; Gifford et aI., 1984; Lenton,

1984; Brun et aI., 1986; Pereto and Beltran, 1987; Aloni et aI., 1988; Estruch et

aI., 1989). Source limitation to overall photosynthate partitioning does exist in

some cases. However, sink demand may also exert control over export from

the leaf (Pereto and Beltran, 1987; reviews by Gifford and Evans, 1981; Gifford

et al. 1984; Ho, 1988; Patrick, 1988). There is appreciable evidence that

phloem translocation from source to sink is not in itself a limiting factor in

photosynthate partitioning (Gifford and Evans, 1981; Gifford et aI., 1984; Delrot

and Bonnemain, 1985; Hanson and Kenny, 1985). Thus, the potential may

exist for improvements in the partitioning of photosynthates from source to sink

by the manipulation of sink strength (Gifford and Evans, 1981; Gifford et aI.,

1984; Kursanov, 1984; Wolswinkel, 1985b; Wyse, 1986; Ho, 1988; Patrick,

1988).

23

The pathways and mechanisms which regulate photosynthate

unloading from the phloem for utilization and/or storage by sinks are as

diverse as the types of tissues and organs that are classified as sinks. It is clear

from a review of the literature that there are many properties of sinks that

require attention before the regulation of sink strength may be understood.

Investigation of the degree of apoplastic versus symplastic phloem unloading

within sink organs, combined with an understanding of the compartmentation

and mechanisms of photosynthate transport within sink cells and tissues are

central to the development of working models for sink processes. The

regulation of photosynthate import, unloading and storage in many sink tissues

by turgor-sensitive mechanisms may provide a basis for the coordination of

phloem import (and unloading) with photosynthate accumulation in sink

organs (Lang, 1983; Wolswinkel, 1985b; Lang and Thorpe, 1986; Ho, 1988;

Patrick, 1988). In particular, the concentration of solutes in the apoplast of sink

tissues may be a key control point for the turgor regulation of phloem

unloading and photosynthate accumulation in sinks (Wolswinkel, 1985b; Lang

and Thorpe, 1986; Ho, 1988; Patrick, 1988). The observation that phloem

loading is enhanced by low cell turgor in source leaves (Daie, 1987a; Aloni et

aI., 1988; Estruch et aI., 1989) may provide a mechanism by which export from

the leaf is coordinated with sink demand, because the turgor differential from

source to sink may be regulated by turgor-sensitive unloading and uptake

mechanisms in sink tissues (Gifford et aI., 1984; Wolswinkel, 1985b; Lang and

Thorpe, 1986; Wolswinkel and Ammerlaan, 1986, 1988; Daie, 1987a; Ho, 1988;

Patrick, 1988). It has been proposed that a high concentration of solutes in the

apoplast of sink tissues is important for the maintenance of low cell turgors

which could support high rates of phloem import by stimulation of mass flow

24

(WolswinkeI1985b; Wolswinkel and Ammerlaan, 1988). This hypothesis may

agree with studies made on several sinks, but many sinks have not been

studied with enough detail to determine whether Wolswinkel's hypothesis is

valid for all sinks.

PHOTOSYNTHATE PARTITIONING DURING THE REPRODUCTIVE

GROWTH OF LEGUMES

The analysis of proven genetic associations between various

characteristics of photosynthate partitioning (photosynthetic rate, leaf export

rate, vegetative development, reproductive development, seed size, seed fill

duration and rate, and harvest index) and seed yield may improve our

understanding of the genetic and physiological determinants of yield in grain

legumes. Study of the interactions between the parameters of seed yield (seed

size, number, fill rate, and fill duration) and experimentally or endogenously

modified patterns of whole plant photosynthate partitioning during reproductive

growth may also be useful for the determination of potential limitations to seed

yield.

Genetic Associations with Seed Yield

Although there is no dependable evidence for a genetic association

between seed yield and photosynthetic rate on an individual leaf basis (Peet et

al., 1977), whole canopy photosynthesis during the period of reproductive

growth appears to be genetically related to seed yield in determinate soybean

cultivars (Wells et aI., 1982) and breeding lines (Ashley and Boerma, 1989).

Phloem transport capacity did not appear to limit assimilate export from

soybean leaves when evaluated on a genetic basis (Hanson and Kenny, 1985).

In addition, the rate of node development before flowering has not been shown

25

to be associated with seed yield (Dunphy et aI., 1979). Genotypes with late

flowering and long seed fill periods appear to have higher seed yields, though

there are clear exceptions to this property (Dunphy et aI., 1979; Hanson, 1985;

Smith et aI., 1988). The selection of higher yielding soybean genotypes based

on seed fill period may be quite difficult in populations segregating for

determinate/indeterminate growth habit (pfeiffer and Egli, 1988), and selection

is also complicated by significant genotype X year interactions (Egli et aI.,

1984).

The duration of the seed fill period is more closely associated with seed

yield than is the rate of seed fill (Egli et aI., 1978; Smith et aI., 1988). However,

seed fill rate is genetically associated with seed size, and seed size is under

genetic control (Egli et aI., 1978). Seed fill rate (and thus seed size) may be

determined by cotyledon cell number, which is under genetic control by the

embryo (Egli et aI., 1981). There is no conclusive evidence for a genetic

association between seed size and seed yield, even though seed size is

genetically determined (Egli et aI., 1978; Hanson, 1988; Smith et aI., 1988). This

has led to the hypothesis that seed number per plant is regulated by the net

production of photosynthate, so that the total seed mass (the product of seed

size and seed number, or yield) may be coordinated with the total production

of photosynthate (Egli et aI., 1978). The observation that harvest index appears

to be genetically determined in soybean may support the hypothesis that seed

yield is coordinated with the total production of photosynthate within a given

genotype (Spaeth et al., 1984). Compensation effects between seed size and

seed number varied between soybean plants in field situations, and the effects

of compensation were only consistent when plants with the same total biomass

were compared (Spaeth and Sinclair, 1984). Thus, there was no evidence that

26

selection for seed size or seed number would be beneficial if selected

independently (Spaeth and Sinclair, 1984). This does not rule out the possibility

for the improvement of seed yield (i.e. harvest index) by the combined

selection for seed number, seed size, or seed fill rate, although it does indicate

that selection for total seed yield is still the most reliable method (but perhaps

the least heritable) for crop improvement through plant breeding.

Summary. The genetic determinants of seed yield in soybean have

only been studied on the most rudimentary level, and only direct selection for

vegetative node development, leaf photosynthetic rate, and leaf transport

capacity may be completely ruled out as methods for crop improvement. The

rate and duration of seed fill, combined with seed size and total seed number

are important to seed yield, but selection for these properties on an individual

basis does not readily lead to improvements in seed yield. Whole canopy

photosynthesis and harvest index may be useful selection criteria, but these

criteria are less reliable and more demanding than direct selection for seed

yield.

Patterns of Whole Plant Photosynthate Partitioning

Sources and sinks during reproductive growth. Seeds and pods are

very strong sinks for photosynthate, and the development of these structures is

associated with major decreases in the partitioning of carbon to the vegetative

parts of legume plants (Kouchi and Yoneyama, 1984a; Yamagata et al., 1987;

Geiger and Shieh, 1988). During the initial period of legume fruit development,

the pod is the primary sink for photosynthate, and pod growth is rapid (Walbot

et aI., 1972; Hsu, 1979; Thorne, 1979; Yamagata et aI., 1987; Geiger et aI.,

1989). After the period of rapid pod expansion, the seeds rapidly become the

greatest sink for photosynthates within the plant, and seed fill is rapid (Walbot

27

et aI., 1972; Hsu, 1979; Yamagata et aI., 1987; Geiger et aI., 1989). The majority

of the carbon stored in legume seeds during the seed fill period is the result of

the direct transport of photosynthates from the leaves to the seeds (Lucas et

aI., 1976; Yamagata et aI., 1987; Geiger and Shieh, 1988). In addition, a

significant proportion of the carbon exported from leaves is stored temporarily

(days) in the stems and pod walls for later remobilization and export to the

seeds (Flinn and Pate, 1970; Thorne, 1979; Fader and Koller, 1985; Peoples et

aI., 1985; Yamagata et aI., 1987; Geiger and Shieh, 1988). Although legume

pods are photosynthetically active, the net production of photosynthates by the

pod does not contribute much to the carbon accumulation of seeds (Flinn and

Pate, 1970; Oliker et aI., 1978; Fader and Koller, 1985; Peoples et aI., 1985).

Source/sink interactions. Low intensity defoliation and pod removal

treatments applied at the end of the flowering period in soybeans did not

significantly modify seed growth rates, which indicates that seed growth rates

were not limited by assimilate supply, perhaps due to feedback control by

seeds over assimilate export from leaves, or due to the buffering effect of

storage in stems versus seeds (Egli and Leggett, 1976). During linear seed fill,

high intensity defoliation was used to demonstrate that soybean seed growth

rates were slowed by drastically limited photosynthate production, although the

duration of seed fill was increased and seed size was unaffected (Egli et aI.,

1985). When the assimilate supply to the seeds was greatly enhanced by

intensive pod removal in the same series of experiments, seed fill rate and seed

size were increased, and the duration of the fill period was lengthened (Egli et

aI., 1985). When shading treatments were used to vary total photosynthate

production during flowering and fruit set, total plant size and growth rate were

modified, but the partitioning between vegetative and reproductive growth was

28

relatively constant, which indicates that seed numbers, which were altered by

the shading treatments, were determined by photosynthate supply (Egli, 1988).

The combined results of the experiments described above indicate that seed fill

rate, seed fill duration, and total seed number can be modified by assimilate

supply, although changes in assimilate supply and demand appear to be highly

buffered, in part by the temporary or long-term storage of photosynthates in

stems (Egli et aI., 1985; Egli, 1988). The mechanisms which regulate seed

number in legumes are unknown, and, even though seed abortion appears to

be under maternal control in pea (Briggs et aI., 1987), the role of assimilate

supply in reproductive abortion is unclear (Heitholt et aI., 1986).

When light and C02 enrichment treatments were temporarily or

continuously applied to soybean plants, the total number and weight of the

pods was increased, and, in plants with temporary enhancement of

photosynthesis, the partitioning of starch and sucrose within the leaf and the

partitioning of photosynthate to the stems was altered relative to the

continuous treatment (Carlson and Brun, 1985). The experiments described

above have been interpreted as evidence for the modification of photosynthate

partitioning within the leaf, and between vegetative and reproductive sinks by

changes in sink demand (Carlson and Brun, 1985). Severe shading treatments

(whole plant excluding one leaf) increased the photosynthetic capacity of the

unshaded leaf in reproductive soybeans, which may indicate that increasing

sink demand increases net photosynthetic capacity (Thorne and Koller, 1974).

The results presented above support the hypothesis that sink demand, which

is primarily accounted for by the pods and seeds, may influence the production

and partitioning of photosynthate within reproductive legumes.

29

The rate of seed fill was relatively unaffected in water stressed soybean

plants, although the length of the fill period was shortened by water stress

(Meckel et aI., 1984). Continuous short days applied to soybean plants after

flowering increased the dry matter accumulation of the fruits (pods + seeds)

versus vegetative tissues, which may indicate that the rate of seed fill responds

to photoperiod controls (Thomas and Raper, 1976). Carlson and Brun (1984)

demonstrated that shortened photoperiods led to greater photosynthate

partitioning to the seeds at the expense of pod growth, and it appeared that

photoperiod effects on starch accumulation in the leaf that were observed in

vegetative plants, were overridden by sink demand in reproductive soybeans. It

is clear from the results presented above, that environmental factors such as

daylength and water stress may alter photosynthate partitioning to the seeds,

although seed demand for photosynthates may modify the response of the

plant to these conditions.

Conclusions

Although a great deal of attention has been focused on the genetic

factors limiting seed yield, there are currently no clearly selectable parameters

that can be used for the genetic improvement of seed yield in grain legumes

(except for seed yield itself). The role of photosynthate supply in limiting seed

yield is also unclear, although it appears that drastic changes in photosynthate

production do affect the parameters of seed yield. The potential for crop

improvement by selection for whole canopy photosynthesis during the period

of reproductive growth may exist, but it is apparent that this must be coupled

with some type of selection for harvest index. The rate and duration of seed fill,

seed size and total seed number are all important parameters for the study of

seed yield, but the relative importance of each of these factors in producing

30

seed yield is ambiguous. The difficulty in establishing the relative importance of

limiting factors during seed growth suggests that knowledge of the

physiological mechanisms which regulate seed fill in legumes may be useful for

the improvement of seed yield. In addition, sink demand by the developing

pods and seeds of legumes may alter whole plant photosynthate partitioning

patterns and may thus be critical to an understanding of the factors which limit

the net production and distribution of photosynthates to the seeds.

PHOTOSYNTHATE PARTITIONING WITHIN DEVELOPING LEGUME

SEEDS

The Growth and Development of Legume Seeds

The rate of dry matter accumulation in developing legume seeds follows

a diauxic pattern, because rapid seed growth appears to be concentrated in

two bursts of seed filling interspersed with a short period of slow growth (Carr

and Skene, 1960, Walbot et aI., 1972; Hsu, 1979; Geiger et aI., 1989). The

majority of seed dry weight accumulation occurs during the second burst of

rapid growth following the lag period (Walbot et aI., 1972; Geiger et aI., 1989).

Cell division in the developing embryo is rapid during early seed development

(until the early cotyledon stage) and the period of rapid cell division correlates

with the initial burst of dry matter accumulation (review by Geiger et aI., 1989).

Following the period of rapid cell division, cell division ceases (review by Geiger

et aI., 1989). The seed coat grows relatively rapidly during the period of rapid

cell division in the embryo, but after this point the embryo increases in mass at

a far higher rate than the seed coat, which ultimately leads to the concentration

of the greatest proportion of total seed mass in the cotyledons (Hsu, 1979;

31

Geiger et aI., 1989). Developing seeds are capable of germinating (Le. mature)

towards the end of the rapid seed fill period, but before the cotyledons turn

yellow and begin to dry down (Walbot et al., 1972).

Hormones and seed growth. Abscisic acid (ABA) stimulates the

growth and protein accumulation of developing soybean embryos during the

early periods of seed development, but suppresses the growth of the embryo

during the middle of the seed fill period (Ackerson, 1984). Thus, the stimulatory

effects of ABA may be the result of the enhancement of cell division (Ackerson,

1984). This may be important to the genetic determination of final seed size,

because seed size and seed growth rate appear to depend on the number of

cells in the cotyledons (Egli et aI., 1981; Ackerson, 1984). The ABA content of

developing soybean seeds was genetically correlated with in situ seed growth

rates and in vitro sucrose uptake rates by developing cotyledons (Schussler et

aI., 1984). An association was also found between large seed size and a high

ABA content in the seed coats (Schussler et aI., 1984). This suggests that ABA

may simultaneously stimulate sucrose release from the seed coat and promote

the growth of the cotyledons (Schussler et aI., 1984). ABA concentrations are

highest in developing soybean seeds during the periods of highest growth, and

are usually higher in the seed coat than in other parts of the seed (or pod)

(Lopez et aI., 1989). This may provide additional evidence for the involvement

of ABA in sucrose unloading from the seed coat (Lopez et aI., 1989). The

partitioning of ABA between the seed coat and the embryo may be regulated

by ABA movement to cells with relatively high pH values, because the

distribution of ABA in the seeds of Phaseolus vulgaris was correlated with a pH

gradient between the cells of the embryo and the cells of the seed coat

(Le Page-Degivry et aI., 1989).

32

The results presented above indicate that ABA is important to legume

seed growth and development, although the potential for growth enhancement

by ABA treatments may be limited. For example, seed injection studies

demonstrated that the addition of ABA and 6-benzylaminopurine (BAP) to

developing Phaseolus vulgaris seeds increased the rate of dry matter

accumulation in situ, but only under conditions of severe defoliation (Clifford

et aI., 1987). The effects of ABA in enhancing the release of sucrose by the

seed coats will be described in greater detail later in this chapter. ABA appears

to be produced in the leaves and transported to the seeds (Brun et aI., 1986;

Le Page-Degivry et aI., 1989).

Seed Coat Structure and the Pathway of Phloem Unloading

The seed coat and embryonic tissues are symplastically separated in

developing legume seeds, so that photosynthates must be unloaded to the

seed coat apoplast before diffusive movement to the embryo (axis +

cotyledons) for uptake (Thorne, 1985; Murray, 1987). Anatomic and

ultrastructural investigation of symplastic continuity in the seed coats of

Phaseolus vulgaris (common bean) demonstrated that sufficient

plasmodesmatal frequency was present to support the symplastic movement

of photosynthates throughout the tissues of the seed coat before efflux to the

apoplast (the symplastic unloading route hypothesis) (Offler and Patrick, 1984).

Less detailed anatomical studies of pea (Pisum sativum) (Hardham, 1976),

soybean (Glycine max) (Thorne, 1981), and broad bean (Vicia faba) (Offler et

aI., 1989) have also demonstrated the potential for some degree of symplastic

phloem unloading within the seed coat. The possibility also exists that

photosynthates are unloaded directly from the phloem to the seed coat

apoplast, without passage through the symplast of other cells of the seed coat

33

(the apoplastic unloading route hypothesis; Wolswinkel, 1987b). Although the

greatest body of evidence supports the symplastic unloading route hypothesis

(Patrick and McDonald, 1980; Thorne, 1981; Hsu et aI., 1984; Offler and

Patrick, 1984; Murphy, 1986; Murray, 1987; Grusak and Minchin, 1988; Offler

et aI., 1989), there has been no definitive confirmation of this pathway

(Wolswinkel, 1987b). The degree of apoplastic versus symplastic phloem

unloading may depend on legume species.

In developing Vicia faba seeds, the transfer cells which line the inner

surface of the seed coat may facilitate photosynthate release to the apoplast,

and transfer cells covering the outer surface of the cotyledons may aid in the

uptake of photosynthates from the apoplast (Offler et aI., 1989). In common

bean, photosynthate release to the apoplast of the seed coat may occur in

thin-walled aerenchymatic branch parenchyma at the inner surface of the seed

coat (yeung, 1983; Offler and Patrick, 1984), and the thickness of this layer is

at a maximum during the period of greatest seed fill rate (yeung, 1983; Hughes

and Swanson, 1985). The structure of soybean seed coats is somewhat similar

to that of common bean in that there is a thin-walled aerenchymatic layer at the

inner surface of the seed coat, although Thorne (1981) has speculated that

efflux to apoplast occurs across the plasmalemmae of the vascular

parenchyma. In soybean, the development of a highly specialized pitjantipit

structure may facilitate some degree of transport between the seed coat and

cotyledons via secretory vesicles, but the evidence for this pathway is limited to

microscopic observations and the physiological significance of this pathway is

unknown (Yaklich et al., 1986). There is no evidence in the literature for the

existence of transfer cells at the surface of the cotyledons in pea, common

bean, or soybean.

34

Sucrose is the major photosynthate transported to developing legume

seeds on a mass basis, and there is no inversion of sucrose between

unloading from the phloem and uptake by the embryo (reviews by Thorne,

1985; Murray, 1987). Significant metabolism of imported amino acids and

ureides may occur in legume seed coats prior to efflux to the apoplast, and the

primary amino acids released from cowpea and soybean seed coats are

asparagine and glutamine (review by Murray, 1987). The metabolic processing

of nitrogenous solutes by the seed coat may provide evidence for a symplastic

route of phloem unloading (Hsu et aI., 1984; Murray, 1987).

The Concentration of Solutes in the Apoplast

The concentration of sucrose in the apoplast between the seed coat

and embryo is quite high (100 to 200 mM) during the period of greatest seed fill

in soybean (Gifford and Thorne, 1985) and common bean (Patrick and

McDonald, 1980; Patrick, 1981). The total amounts of sucrose, organic acids,

amino acids and ions in the liquid endosperm of common bean changed

during seed development, from 500 mOsm at the globular heart stage to

700 mOsm at the late cotyledon stages (Smith, 1973). Sugar concentrations

(primarily sucrose, glucose and fructose) increased between the early and late

cotyledon stages from a total of 105 to 170 mM respectively. The analytical

methods used by Smith (1973) were not extremely precise, and thus caution

must be used in interpreting the data presented above. In general, the data that

has been presented by many researchers supports a high concentration of

solutes (in the vicinity of 350 mOsm) in the apoplast of legume seed coats

(Thorne 1985a; Wolswinkel, 1985b).

Water Relations of the Seed Coat and Cotyledons

In soybean seed coats during the period from the early cotyledon stage

to maturity, the osmotic potential increased slightly from -1.6 MPa to -1.4 MPa,

35

and then declined rapidly near the end of the seed fill period to -1.9 M Pa (Saab

and Obendorf, 1989). The pressure potential ofthe soybean seed coat

decreased during the same period from +0.6 MPa to +0.2 MPa (Saab and

Obendorf, 1989). The solute and pressure potentials of soybean cotyledons

remained fairly constant from the early cotyledon stage to the late cotyledon

stage at -1.2 and +0.5 MPa, respectively (Saab and Obendorf, 1989). The

osmotic and pressure potentials of the cotyledons decreased dramatically near

maturity to -2.4 and -0.1 MPa respectively (Saab and Obendorf, 1989). The

decline in the turgor of seed tissues near maturity is probably due to

desiccation prior to dormancy (Saab and Obendorf, 1989). The measurements

of Saab and Obendorf (1989) demonstrate that the turgor of soybean seed

coats was lower than the turgor of the cotyledons after the early cotyledon

stage of development.

The osmotic potential of Phaseolus seed coats remained fairly constant

from the early heart stage to maturity at a value of -1.0 MPa (Yeung and Brown,

1982). The pressure potential (turgor) of Phaseolus seed coats for the period

from the early heart stage to maturity peaked at a value of +0.3 MPa during the

middle of this period and then declined back to its original value of +0.1 MPa

(Yeung and Brown, 1982). The osmotic potential of the cotyledons for the

period between the early heart stage and maturity decreased from -1.0 MPa to

-2.2 MPa, while the pressure potential of the cotyledons followed a different

trend during the same period, increasing to a value of +0.8 MPa, from its initial

and final value of +0.3 MPa and then declining to +0.5 MPa at maturity (Yeung

and Brown, 1982). The osmotic potential of the liquid endosperm in seeds of

common bean could be measured over the period between the early heart and

the early cotyledon stages of development, and was found to be relatively

36

constant at approximately -1.1 MPa (Yeung and Brown, 1982). Thus, the turgor

of Phaseolus seed coats is maintained at fairly low levels when compared with

those of the cotyledons, and this may be due in part to a low solute potential in

the apoplast (Le. the liquid endosperm).

In contrast with the results of Yeung and Brown (1982) and Saab and

Obendorf (1989), the osmotic potential of the seed coat was more negative

than that of the embryonic tissues until near maturity in round-seeded peas

(the RR genotype), and the same was true of wrinkled-seeded peas (the rr

genotype) until the middle of the seed fill period (Wang et aI., 1987). The

osmotic potential of the liquid endosperm (when present) was always less

negative than the osmotic potential of the seed coat or cotyledons

(approximately -1.4 MPa; Wang et aI., 1987). The pressure potential of seed

tissues was not measured by Wang et al. (1987), and thus the turgor of seed

coats and cotyledons could not be compared.

The results summarized above indicate that changes in the osmotic

environment of seed tissues may be important to various developmental

processes including seed maturation and desiccation (Yeung and Brown,

1982; Wang et aI., 1987; Saab and Obendorf, 1989). In addition, the relatively

low turgor of developing seed coats and the low osmotic potential of the liquid

endosperm appear to support Wolswinkel's (1985b) hypothesis that low turgor

in the seed coat may be important to the enhancement of phloem transport to

this tissue.

Sucrose Uptake by the Developing Embryo

Sucrose uptake by developing soybean cotyledons is at least partially

mediated by a sucrose/proton symport, which is driven by the proton

electrochemical potential gradient produced by an electrogenic H + -ATPase in

the plasmalemma of cotyledon cells (Lichtner and Spanswick, 1981a, 1981b;

37

Lin, 1985). Sucrose uptake by developing soybean cotyledons may be

resolved into at least two components: a saturable phase and a linear phase

(Uchtner and Spanswick, 1981b; Lin, 1985). There is also evidence for an

energy-dependent mechanism of sucrose uptake in the cotyledons of

Phaseolus vulgaris, and saturable and linear components of uptake were also

observed in this system (Patrick, 1981). At the concentrations of sucrose that

have been estimated for the apoplast in vivo, the linear component of sucrose

uptake appears to dominate saturable uptake in soybeans (Lichtner and

Spanswick, 1981b; Thorne, 1982b; Lin, 1985; Gifford and Thorne, 1985). The

linear component of sucrose uptake appears to consist of a passive

component (diffusion) and a component that may be mediated by

sucrose/proton cotransport (Lin, 1985). Thus it appears that both saturable

and linear components may be mediated to some degree by sucrose/proton

cotransport (Lin, 1985).

The saturation of active carrier-mediated sucrose uptake at in vivo

sucrose concentrations indicates that carbon accumulation may be limited by

the rate of sucrose uptake by the cotyledons of soybean (Lichtner and

Spanswick, 1981b) and Phaseolus vulgaris (Patrick, 1981). Sucrose appears to

move within the apoplast of the cotyledons by diffusion, and this may also limit

the rate of sucrose uptake by the cotyledons (Patrick and McDonald, 1980).

Thus, there may be a sucrose concentration gradient within the cotyledons

such that sucrose uptake at the surface of the cotyledons may be primarily

passive (high sucrose concentration) while sucrose uptake in the interior of the

cotyledons (low sucrose concentration) may require active uptake (Gifford and

Thorne, 1985). The possibility that carbohydrate metabolism may also regulate

carbon accumulation in developing legume seeds by lowering the

38

concentration of sucrose in the symplast of the cotyledons has not been

examined in detail (Sung et aL, 1989; Thorne, 1982b). Sucrose uptake by

developing cotyledons of legumes may be increased by low cell turgor,

although the evidence for this hypothesis was not obtained by direct

measurements of sucrose uptake (Wolswinkel and Ammerlaan, 1986;

Wolswinkel et aL, 1986).

The uptake of amino acids by developing pea and soybean embryos

(axis + cotyledons) appears to be enhanced near the end of the seed fill

period by an increase in the activity of active (saturable) uptake mechanisms

(De Ruiter et aL, 1984; Vernooy et aL, 1986; Cornish and Spanswick, 1987;

Lanfermeijer, 1987; Lanfermeijer et aL, 1989). The saturable component of

glutamine and glutamate uptake in developing soybean embryos appears to

be mediated by an amino acid/H + cotransport mechanism (Bennett and

Spanswick, 1983; Cornish and Spanswick, 1987). As observed for sucrose

uptake, both saturable and linear components of uptake are important to

amino acid uptake in developing embryos of pea (Cornish and Spanswick,

1987; Lanfermeijer, 1987; Lanfermeijer et aL, 1989) and soybean (Bennett and

Spanswick, 1983; Cornish and Spanswick, 1987). The uptake of valine by

developing soybean embryos was decreased by low cell turgor, a result that

may indicate that the turgor of the embryo (which is influenced by the

concentration of solutes in the apoplast) may regulate amino acid uptake in this

tissue (Guldan and Brun, 1987).

The Empty Seed Coat Technique

The empty seed coat technique allows the measurement of

photosynthate unloading from developing legume seed coats following the

surgical removal of the embryo through an incision in the seed coat, so that a

perfusion solution (or agar trap) may be introduced (Thorne and Rainbird,

39

1983; Wolswinkel and Amerlaan, 1983; Patrick, 1983). Materials released to the

perfusion solution by the seed coat may thus be collected for analysis, and the

effects of various treatments on the kinetics of solute or tracer release from the

seed coat may be measured. The empty seed coat technique has been used

by many researchers to study the mechanisms governing photosynthate

unloading in developing legume seed coats (reviews: Thorne, 1985; Murray,

1987; Wolswinkel, 1985b, 1988). An important principle behind the use of the

empty seed coat technique is that the manipulations involved in this procedure

leave "the maternal, assimilate-delivering tissue intact and functioning"

(Wolswinkel, 1988). However, the mechanisms of phloem import and

unloading may be disrupted by the surgical modification and perfusion of

seeds as has recently been reassessed by Minchin and Thorpe (1989).

The complex compartmentation of substances within seed coats

complicates the analysis of materials eluted during seed coat perfusion,

because materials may be eluted from the apoplast or the symplast, and

photosynthates imported to the seed coat via the phloem may be released

either directly from the phloem, or after passage through the symplast. In order

to clarify discussion of photosynthate movement into and out of perfused seed

coats, the following terms will be used, as proposed by Grusak and Minchin

(1988); phloem import: the movement of photosynthates into the seed coat

via the phloem; phloem unloading: the movement of assimilates from the

phloem to the apoplast or the symplast of the seed coat; and seed coat

unloading: the movement of photosynthates from the seed coat symplast to

the apoplast.

Effects of EDTA on photosynthate unloading from perfused seed

coats. Chelating agents such as EDTA and EGTA have been used for the

collection of phloem sap from cut petioles and other tissues, because these

40

chelators appear to "unclog" the phloem, leading to high rates of phloem

exudation over prolonged periods (Groussol et aI., 1986). Groussol et al.

(1986) measured the sucrose and hexose content of petiole exudates in order

to identify the source of exuded sugars, based on the principle that pure

phloem sap contains much more sucrose than hexose. At higher

concentrations of EDTA (and EGTA) there was an increase in the hexose

content of petiole exudates, and thus higher concentrations of chelating agents

were said to produce a general "leakiness" of the cortical cells in exuding

petioles. At 15 mM EDTA (the concentration used in chapter 4 of the current

study and in Wolswinkel, 1987a) 60% of the sugar in petiole exudates was

sucrose. Although 7% of the hexoses were produced by sucrose inversion

following exudation, at least 33% of the sugar released by treatment with

15 mM EDTA was due to general leakage from cortical tissues. Groussol et al.

(1986) did not account for sucrose leakage from cortical cells, and thus the

general leakiness induced by EDTA (and EGTA) treatments may have been

underestimated.

Thorne and Rainbird (1983) measured the effects of EGTA and EDT A on

photosynthate efflux from soybean seed coats into agar traps using the empty

seed coat technique. It was found that 15 mM EGTA stimulated the efflux of

pulse-labeled 14C-photosynthate efflux into agar traps, apparently at the

expense of 14C retention in the seed coat. The same effect was observed with

EDTA, although smaller in magnitude. EGTA-induced increases in 14C_

photosynthate efflux to agar traps were said to result from massive phloem

leakage, and a general increase in membrane permeability. Although there was

no apparent increase in 14C import to the entire modified seed (seed coat +

agar), EGTA was also said to promote phloem import to the seed.

41

Hanson (1986a) studied the effects of 2 versus 20 mM EDTA on sucrose

efflux from "empty" soybean seed coats into a trapping solution and found that

20 mM EDTA enhanced sucrose efflux more than 2 mM EDT A, and also

demonstrated that the increase in efflux caused by EDTA became greater over

time relative to a solution without EDT A. Treatment with 20 mM EDTA

produced a sucrose efflux that was not proportional to the area of the seed

coat, while sucrose efflux from seed coats treated with 2 mM EDTA was more

dependent on seed coat area. When seed coats were treated with 200 mM

sorbitol and 2.5 mM EDTA, sucrose efflux demonstrated an approximately

linear dependence on seed coat area. The dependence of sucrose efflux on

seed coat area was used as a measure of whether efflux was related to

sucrose import to the seed coat via the phloem, although the amount of

sucrose present in the seed coat would also be expected to depend on seed

coat area, regardless of phloem import. Although EDTA was used to facilitate

long-term sucrose efflux measurements, it appears that a general leakiness

may be induced by EDTA, and the results of these experiments may not

accurately reflect phloem import and unloading of sucrose from seed coats.

Wolswinkel (1987a) treated attached Vicia faba seed coats with 400 mM

mannitol and 15 mM EDTA and found that the efflux of unlabeled sugars

(sucrose + glucose) and amino acids was increased over efflux from a

400 mM mannitol control. As observed previously by Hanson (1986a), the

efflux enhancing effect of EDTA increased over time. The same effect of EDTA

was observed with excised seed coats, although the effect was less

pronounced. On the basis of these experiments, Wolswinkel (1987a)

concluded that both phloem import and photosynthate release were stimulated

by EDTA, although the effect of EDTA on net phloem import was not directly

42

demonstrated. In summary, the effects of EDTA on sugar release from

perfused seed coats appear to be the combined result of a non-specific

increase in the leakiness of seed coat membranes, and a specific stimulation of

photosynthate release from the phloem. The role of EDTA in stimulating

phloem import to the seed coat remains speculative.

Turgor-sensitive phloem import and seed coat unloading. Higher

concentrations of osmoticum in the perfusion solution (up to 600 mM)

increased the efflux of pulse-labeled photosynthates and unlabeled sugars

(and amino acids) from seed coats attached to the plant via the funiculus

(Wolswinkel and Ammerlaan, 1984, 1985a, 1986; Wolswinkel et al. 1986;

Minchin and McNaughton, 1986; Ellis and Spanswick; 1987; Grusak and

Minchin, 1988). In addition, lower osmoticum concentrations (100 mM, or

solutions without osmoticum) in the perfusion solution increased the efflux of

14C-photosynthates from excised seed coats (Patrick, 1983, 1984; Patrick et

al. 1986; Wolswinkel and Ammerlaan, 1986; Wolswinkel et aI., 1986; Grusak

and Minchin, 1988). Experiments with excised versus attached seed coats may

yield different results because the net efflux of photosynthates from attached

seed coats may be dominated by net phloem import, while the efflux of

photosynthates from excised seed coats may be determined primarily by

mechanisms that regulate efflux from the seed coat itself (Wolswinkel et aI.,

1986; Grusak and Minchin, 1988). The enhancement of photosynthate efflux

from attached seed coats by high concentrations of osmoticum may be

explained by an increase in phloem import due to low cell turgor in the seed

coat (Wolswinkel and Ammerlaan, 1984, 1985a, 1986, 1988; Wolswinkel et aI.,

1986; Minchin and McNaughton, 1986; Ellis and Spanswick, 1987; Grusak and

Minchin, 1988). Osmoticum treatments did not induce changes in seed coat

43

membrane permeability as determined from 14C-mannitol uptake experiments,

and thus the increase in photosynthate efflux observed at low osmoticum

concentrations could be explained by either an increase in efflux through a

turgor-sensitive carrier (Patrick, 1984; Patrick et aI., 1986), or by a decrease in

uptake by a turgor-sensitive apoplastic retrieval mechanism (Wolswinkel and

Ammerlaan, 1986; Wolswinkel et aL 1986). 14C-photosynthate efflux

experiments provide substantial support for the existence of a turgor-sensitive

photosynthate efflux control mechanism in seed coats (Patrick, 1984; Patrick et

aL, 1986). Evidence for a turgor-sensitive apoplastic sugar retrieval mechanism

is circumstantial, as efflux measurements were used to provide data on sugar

uptake rates (Wolswinkel and Ammerlaan, 1986; Wolswinkel et aL 1986).

Although the mechanism responsible for the turgor-regulation of photosynthate

unloading from legume seed coats has not been conclusively determined, the

unloading of photosynthates from the seed coat does appear to be enhanced

by high cell turgor (low concentrations of osmoticum in the apoplast).

The existence of a mechanism which regulates the net release of

photosynthates from the seed coat in response to cell turgor may allow the

coordination of phloem import, seed coat unloading, and photosynthate

uptake by the embryo (Patrick, 1984; Wolswinkel, 1985b, 1988; Patrick et aL,

1986; Wolswinkel et aL, 1986). This is made possible by the interaction of all

these mechanisms with the concentration of solutes (primarily sucrose) at the

apoplastic interface between the seed coat and the embryo. Increases in the

rate of sucrose uptake by the embryo would be expected to decrease the

concentration of sucrose in the apoplast, which would thus stimulate sucrose

release by changing the activity of a turgor-sensitive efflux control mechanism.

Therefore, turgor-sensitive efflux control could be critical in sustaining a high

44

concentration of sucrose in the apoplast, which would maintain a low turgor in

the seed coat and thus stimulate phloem import to the seed (Patrick, 1984;

Wolswinkel and Ammerlaan, 1984, 1988; Wolswinkel, 1985b, 1988; Patrick et

aI., 1986; Wolswinkel et aI., 1986). The observations that the turgor of seed

coats is maintained at a fairly low level during most of the seed fill period

(yeung and Brown, 1982; Saab and Obendorf, 1989), and that the

concentration of solutes in the apoplast is relatively high during the same

period (Smith, 1973; Patrick and McDonald, 1980; Patrick, 1981; Gifford and

Thorne, 1985) may provide evidence that the turgor-regulation of

photosynthate unloading from the seed coat may function as part of a

feedback mechanism that coordinates the rate of sucrose uptake by the

embryo with the rate of phloem import to the seed. The regulation of seed coat

turgor by a turgor-sensitive efflux control mechanism may also help to maintain

a constant turgor differential from source to sink in the phloem which would

thus sustain mass flow from source to sink (Patrick, 1984; Wolswinkel and

Ammerlaan, 1984, 1988; Wolswinkel, 1985b, 1988; Lang and Thorpe, 1986;

Patrick et aI., 1986). The concentration of solutes in the apoplast may change

during seed development (Smith, 1973; Murray, 1987;), and there is evidence

that the optimal osmotic potential for photosynthate unloading from the seed

coat may also change over the developmental period (Wolswinkel and

Ammerlaan, 1986). Thus, the concentration of solutes in the apoplast and its

relationship to the water relations of the seed coat and the embryo may be

important to the long term regulation of the processes that control the rate and

duration of seed fill (Thorne, 1985; Vernooy et aI., 1986; Murray, 1987;

Wolswinkel, 1988).

45

Energetics and carrier-mediation of photosynthate unloading. In

general, the release of pulse-labeled and unlabeled photosynthates (sucrose

and amino acids) from empty seed coats (Patrick, 1983; Thorne and Rainbird,

1983; Wolswinkel and Ammerlaan, 1983; Wolswinkel et aI., 1983; Gifford and

Thorne, 1986) and intact seed coats (Thorne, 1982a) appears to be inhibited

by metabolic inhibitors such as anoxia, low temperature, NaN3, CCCP, KCN,

NEM, FCCP, H202, NaF, NaAs02, and DNP. The observation that

photosynthate unloading from seed coats appears to be sensitive to metabolic

inhibitors may provide evidence of an energy requirement for photosynthate

unloading from the seed coat (Thorne, 1982a; Patrick, 1983; Thorne and

Rainbird, 1983; Wolswinkel and Ammerlaan, 1983; Wolswinkel et aI., 1983;

Gifford and Thorne, 1986; Murray, 1987; Wolswinkel, 1988). Some metabolic

inhibitors appear to restrict phloem import or unloading (FCCP, KCN, NaF,

NaAs02, NEM, DNP) (Thorne and Rainbird, 1983; Wolswinkel and Ammerlaan,

1983; Wolswinkel et aI., 1983; Gifford and Thorne, 1986; Minchin and

McNaughton, 1986; Grusak and Minchin, 1988) while some inhibitors may

restrict symplastic transport within the seed coat (CCCP, NEM) (Patrick, 1983;

Grusak and Minchin, 1988). It is often difficult to determine whether metabolic

inhibitors such as those used in the studies described above act on specific

processes or whether they produce general disruptions of many processes. In

addition, the different methods by which metabolic inhibitors have been applied

to seed coats, and the conflicts between the results of different researchers,

demonstrate that although the effects of metabolic inhibitors on photosynthate

unloading from seed coats are suggestive of an energy requirement for phloem

unloading and/or seed coat unloading, the results of metabolic inhibitor

studies are not conclusive.

46

PCMBS, a non-penetrating sulfhydryl reagent which has been proposed

to inhibit carrier-mediated processes located at the plasmalemma, has been

used to determine whether photosynthate unloading from the seed coat is a

carrier-mediated process (Patrick, 1983, Thorne and Rainbird, 1983;

Wolswinkel and Ammerlaan, 1983, 1986; Wolswinkel et aI., 1983; Wolswinkel,

1985a; Gifford and Thorne, 1986; Minchin and McNaughton, 1986; Patrick et

aI., 1986; Grusak and Minchin, 1988). Most researchers have concluded that

PCMBS produces a reversible (by treatment with OTT or OTE) inhibition of

sucrose efflux from seed coats, and that this observation is evidence for the

carrier-mediation of sucrose unloading from seed coats (Patrick, 1983, Thorne

and Rainbird, 1983; Wolswinkel and Ammerlaan, 1983, 1986; Wolswinkel et aI.,

1983; Gifford and Thorne, 1986; Patrick et aI., 1986). Conflicts over the

pathway of phloem unloading are apparent in the study of PCMBS effects on

photosynthate unloading from seed coats, as some workers claim an inhibition

of phloem unloading to the apoplast (Thorne and Rainbird, 1983; Wolswinkel et

aI., 1983; Gifford and Thorne, 1986), while others claim that unloading from the

interconnected seed coat symplast is the primary site of PCMBS inhibition

(Patrick, 1983; Patrick et aI., 1986). It has also been proposed that an energy­

dependent, carrier-mediated apoplastic retrieval mechanism in seed coats is

inhibited by PCMBS (Wolswinkel, 1985a).

Although there is a great deal of evidence for energy-requiring, carrier­

mediated photosynthate unloading mechanisms in legume seed coats, there is

also a great deal of conflict over the types of energy-requiring, carrier-mediated

mechanisms that are responsible for the regulation of photosynthate

unloading. As a further complication, Minchin and McNaughton (1986) and

Grusak and Minchin (1988) have demonstrated that neither metabolic inhibitors

47

nor PCMBS appear to affect photosynthate unloading from pea seed coats, as

the effects of these treatments were quite slow in their experiments. Patrick et

al. (1986) also observed that turgor-sensitive photosynthate unloading did not

appear to be inhibited by PCMBS, and concluded that this was due to some

type of mechanism which protected the turgor-sensitive carrier from PCMBS.

The potential for a passive, turgor-sensitive, non-carrier-mediated mechanism

of photosynthate efflux control was discussed by Grusak and Minchin (1988),

but the nature of such a mechanism remains elusive. Despite a great deal of

research, energy-dependent carrier-mediated photosynthate unloading

mechanisms are still mysterious, and apparently even more research into the

energetics and carrier-mediation of photosynthate unloading from legume seed

coats is justified. Evidence for passive mechanisms of photosynthate unloading

from seed coats is also sparse.

Is sucrose/H + cotransport involved in photosynthate unloading?

Van Bel and Patrick (1984) have proposed that sucrose unloading from

common bean seed coats is mediated by an outwardly directed

sucrose/proton symport that is not coupled with an electrogenic proton pump.

Although a variety of conclusions were reached as to the effects of fusicoccin,

K+, orthovanadate, and ABA on the efflux of protons and 14C-photosynthates,

the evidence gathered in support of the hypothesis stated above is clearly

circumstantial, as membrane potential measurements were not used in these

experiments. Minchin and McNaughton (1986) demonstrated that 11 C­

photosynthate release from seed coats was increased by treatment with a high

pH solution, and concluded that this was due to an increase in the leakiness of

phloem plasmalemma, and not to a change in the activity of a proton/sucrose

48

cotransport mechanism. There is currently no coherent model for the

involvement of a sucrose/proton cotransport mechanism in the regulation of

photosynthate unloading from legume seed coats, although this possibility is

clearly worthy of further study.

The effects of K + , growth regulators, and other treatments on

photosynthate unloading. There is considerable,evidence for the

enhancement of photosynthate efflux from seed coats by the addition of K + to

the perfusion solution (Van Bel and Patrick, 1984; Wolswinkel and Ammerlaan,

1985b; Clifford et aI., 1986; Patrick, 1987). The mechanism by which K+

enhances photosynthate efflux from seed coats is speculative, although it has

been proposed that K + acts in balancing the charge differential across the

plasmalemma of seed coat cells, and thus stimulates a sucrose/proton

cotransport mechanism (Van Bel and Patrick, 1984; Patrick, 1987). The

stimulatory effect of K+ on photosynthate efflux may be due to changes in the

activity of a sucrose uptake mechanism in the seed coat (Wolswinkel and

Ammerlaan, 1985b) or to a change in the activity of a carrier that directly

controls sucrose efflux (Van Bel and Patrick, 1984; Patrick, 1987).

ABA has been shown to increase the efflux of sucrose (Gifford and

Thorne, 1986), and pulse-labeled photosynthate from perfused seed coats

(Van Bel and Patrick, 1984; Clifford et aI., 1986; Ross et aI., 1987). The efflux­

enhancing effects of ABA were increased by the addition of K + in the

experiments of Clifford et al. (1986), but not in the experiments of Ross et al.

(1987). Indoleacetic acid (IAA) was shown to stimulate sucrose efflux by Gifford

and Thorne (1986), although this was not confirmed by Clifford et al. (1986).

Clifford et al. (1986) studied the effects of many growth regulators and found

that in addition to ABA, 6-benzylaminopurine (BAP, a cytokinin) stimulated the

49

efflux of 14C-photosynthate, although many other growth regulators, including

several gibberellins and NAA did not enhance photosynthate efflux. In addition,

ACC (a substrate for ethylene production) had no measurable effects on

photosynthate efflux (Clifford et aI., 1986). It was also found that the stimulation

of photosynthate efflux caused by BAP was enhanced by K+ (Clifford et aI.,

1986). Fusicoccin, which is a stimulant of electrogenic proton pumping (and a

potent toxin) was shown to enhance the efflux of sucrose from soybean seed

coats (Gifford and Thorne, 1986), while it inhibited the efflux of 14C_

photosynthate from Phaseolus vulgaris seed coats (Van Bel and Patrick, 1984).

The conflicts observed in the effects of various treatments described above

may be dependent on technique, as Gifford and Thorne (1986) used sucrose

efflux measurements with attached seed coats, while Van Bel and Patrick

(1984), and Clifford et al. (1986) measured the efflux of pulse-labeled 14C_

photosynthates from detached seed coats. Although evidence has been

presented for the hypothesis that growth regulators act on photosynthate

unloading from seed coats through the modification of sucrose cotransport

mechanisms (Van Bel and Patrick, 1984; Clifford et aI., 1986; Ross et aI., 1987),

this is clearly speculative. In summary, the results presented above indicate

that ABA and K+ consistently stimulate photosynthate efflux from seed coats,

while a limited amount of evidence is available for the stimulation of

photosynthate efflux from seed coats by IAA, BAP, and fusicoccin.

CONCLUSIONS

The physiological mechanisms which regulate photosynthate

partitioning in reproductive legume plants are numerous, and interact with each

other in such a way that individual mechanisms that limit seed production are

50

difficult to resolve. Although the investigation of individual mechanisms of

photosynthate partitioning by a reductionist approach is required for a detailed

understanding of these mechanisms, it is clear that the improvement of seed

yield in legumes will require the study of whole plant processes. Steady state

labelling with 14C may provide a methodology for the study of photosynthate

partitioning mechanisms on many levels, and may also permit the analysis of

the interactions of multiple processes within whole plants.

The mechanisms which regulate the partitioning of photosynthate within

developing legume seeds may be studied with relative convenience through

the use of the empty seed coat technique. The use of this technique has

already enhanced our understanding of phloem unloading within the seed

coat, although most of the information gathered to date is qualitative in nature.

The combined use of the empty seed coat technique and steady state labelling

may facilitate a quantitative approach to the analysis of photosynthate

unloading in developing legume seed coats, and may also allow the coupling

of empty seed coat studies to the analysis of whole plant photosynthate

partitioning. The involvement of turgor-sensitive mechanisms in the regulation

of photosynthate unloading in developing legume seeds appears to offer the

potential for the regulation of partitioning processes at both the seed and at the

whole plant level. The purpose of this study is to evaluate the quantitative

significance of turgor-sensitive photosynthate unloading in developing legume

seed coats. In addition, steady state labelling methods were developed for use

with the empty seed coat technique so that quantitative methods for the

analysis of photosynthate partitioning will be available for the determination of

limiting factors to the rate of seed fill and ultimately to the seed yield of grain

legumes.

CHAPTER 2

The Use of Steady State Labelling to Study Phloem Transport into

Developing Seeds

INTRODUCTION

The agronomic and biological importance of the processes which

regulate the partitioning of carbon between leaves and developing seeds has

inspired a great deal of research, and many recent studies have focused on the

mechanisms of carbon allocation within developing legume seeds (Thorne,

1985; Murray, 1987; Wolswinkel, 1988). The seed coat and embryo of legume

seeds are not symplastically linked and therefore photosynthates entering the

embryo must be unloaded to the seed coat apoplast before uptake (Thorne,

1985). The route of photosynthate movement from the phloem sieve tubes to

the seed coat apoplast has not been conclusively determined, although the

hypothesis that photosynthates must move symplastically from the phloem

throughout the tissues of the seed coat before unloading to the apoplast (the

symplastic unloading route hypothesis) has received support from a variety of

studies (Patrick and McDonald, 1980; Offler and Patrick, 1984; Murphy, 1986;

Grusak and Minchin, 1988; Offler et aI., 1989). An alternative to the symplastic

unloading route hypothesis is that photosynthates are unloaded directly from

the phloem to the seed coat apoplast, without passage through the symplast of

other cells of the seed coat (the apoplastic unloading route hypothesis)

(Wolswinkel, 1987b).

The study of photosynthate unloading from legume seed coats has

been enhanced by the development of the "empty seed coat technique", which

involves the surgical removal of the embryo through an incision in the seed

coat so that a perfusion solution (or agar trap) can be used to collect materials

51

52

moving out of the seed coat for subsequent analysis (Thorne and Rainbird,

1983; Wolswinkel and Ammerlaan, 1983; Patrick, 1983). Although several

studies have attempted to determine the sources of materials collected by the

perfusion of empty seed coats, there is still confusion as to the quantities of

various photosynthates that are unloaded via the typical unloading

mechanisms of the seed coat in contrast to materials removed due to a simple

washout of the cells within the seed coat (Gifford and Thorne, 1986; Patrick et

aI., 1986; Wolswinkel et aI., 1986). Additional complications in the use of the

empty seed coat technique are that the surgical procedure, or the introduction

of a solution into the seed coat "cup" may disturb phloem import into the seed

coat and/or the process of unloading to the seed coat apoplast (Minchin and

Thorpe, 1989). Early studies using pulse-labelling of leaves (or petioles) found

that the amounts of labeled photosynthates imported to perfused seeds were

similar to intact seeds (Patrick, 1983; Thorne and Rainbird, 1983; Wolswinkel

and Amerlaan, 1983). However, analyses of tracer accumulation in tissues

following pulse-labelling are complicated by constantly changing fluxes of

tracer, and thus the validity of quantitative comparisons between tracer import

by perfused versus intact seeds are questionable (Geiger and Shieh, 1988).

Many problems inherent in pulse-labelling are circumvented by the use

of steady state labelling techniques. Steady state labelling has been used for

the quantitative analysis of various aspects of carbon partitioning in plants,

because the stable tracer fluxes produced by the continuous assimilation of

14C02 at a constant specific activity result in relatively simple tracer kinetics at

the sink (Geiger, 1980; Geiger and Shieh, 1988). The purpose of this study was

to demonstrate that the methodology of steady state labelling is useful in the

quantification of 14C transport into and within the developing seeds of

53

Phaseolus vulgaris. The import of 14C into surgically modified seeds (as

required for the empty seed coat technique) was quantified to determine

whether the fluxes of photosynthate within cut seeds were significantly different

from those within intact seeds.

MATERIALS AND METHODS

Plant material. A determinate red kidney bean cultivar (Phaseolus

vulgaris L., cv. Redkloud, obtained from Dr. Donald Wallace, Department of

Plant Breeding, Cornell University, Ithaca, NY, U.S.A.) was grown in 2 gallon

pots in the greenhouse during the winter and spring of 1988. Supplemental

lighting provided by 750 W mercury halide lamps (200 ",mol photons m-2 s-1

after sundown at mid-canopy height) was used to maintain the photoperiod at

16 h. Plants were fertilized bi-weekly with a dilute solution of Peter's

Professional Water Soluble Fertilizer fY'J. R. Grace & Co., Fogelsville, PA,

U.S.A.) applied to the soil, and plants were regularly sprayed with various

pesticides to control insects. Nodes were cleared of all vegetative lateral buds

on a weekly basis, so that a uniform, determinate growth habit was achieved.

Flowering began at the terminal raceme and top axil (7th node) approximately

33 days after planting. Seeds had ceased filling approximately 34 days after

flowering, as indicated by maximum size and the onset of cotyledon yellowing.

Plants were selected for experiments 15 to 23 days after flowering, at which

point the seeds had reached a fresh weight of 500 to 1000 mg.

Steady state labelling and carbon fixation measurement. Steady

state labelling, of leaves with 14C02 was accomplished using the apparatus

illustrated in Figure 2.1. The materials used in system construction were

aluminum, copper, brass and glass, because these materials are not

54

Figure 2.1 Diagram of steady state labelling apparatus. GM: Geiger-MOiler

detector; PM1,PM2: pressure meters; V1: 2-way valve; V2: 3-way valve; V3-V?:

4-way valves; IRGA: Infra-red gas analyzer. The closed-circuit flow path used

for steady state labelling is as follows: starting at the peristaltic pump, air flows

through V4 into the ballast tank and out again to V4; after V4, air passes

through a particle filter, pressure meter, IRGA, and flow meter to V5, and then

from V5 to V6 to the leaf chamber; air then moves out of the leaf chamber to

V7, and from V? to V6, through the condenser, water trap, and pressure meter

to V3, and finally, from V3 back to the peristaltic pump. C02 stored in a

regulated cylinder enters the system via the low pressure regulator and

solenoid valve under computer control. Valve V2 allows the selection of either

labelled or unlabelled C02' For C02 removal from the system, valve V3 is reset

to allow air flow through the C02 trap (NaOH solution), and the cold finger (at

O°C) prevents vapor flow into the system. Manipulation of valves V5-V7

facilitates the exclusion of the leaf chamber from the rest of the system and/or

the flushing of the system with room air. Switching V4 causes air to be diverted

from the ballast tank, allowing rapid IRGA calibration with external standard

gases. Humidity is maintained at a constant dewpoint using the temperature­

controlled condenser and water trap, and valve V1 facilitates drainage of

trapped water (from transpiration) at the end of an experiment.

55

COMPUTER

AID ffim DIGITAL SOLENOID -

CONVERTER ~ DRIVER ,--

0 OUT 0 0

FI LTER I -

f AMPLIFIER r---- FLOW r----- METER ,-.-

:FILTER J==(S)= -

'""="" I R GA r-Ol--r-- t--

'-=-

V4 PM2 , L b ~ p

t FAN LEAF CHAMBER

9~Qt ~~ -r---1..-

PERI-BALLAST 0 t STALTIC

PUMP TANK FAN GM L :r r: :r r: h

-E--L

t OUT CJ L ~ L.:1 p-

~ PUT V7 V6 ~ IN V5

--k h PUTt

V3L. ~

~ -~ I - K;;;;> t - COLD § CON-• .

DENSER ·CO • • 2 FINGER 0

o TRAP '-' ~

...- ~f- WATER / "- \::J- TRAP

~ PMI 101 VI

c::f7J 101 I V2 I

0 r:: Cl ® ~I SOLENOID !-----I ~ I VALVE

COLD 14CO

LOW CO

2 PRESSURE 2 REGULATOR

56

phytotoxiC (Geiger and Shieh, 1988). The peristaltic pump (Varistaltic advanced

model, Manostat, New York, NY, U.S.A.) used to circulate air within the system

was equipped with non-toxic silicone rubber tubing (J-6411-71, Cole-Parmer

Instrument Co., Chicago, IL, U.S.A.). A 13 L "ballast tank" was added to the

system to increase its internal volume so that the rate of C02 concentration

change would be decreased. C02 measurements were made using an infrared

gas analyzer (LIRA model 303, Mine Safety Co., Pittsburgh, PA, U.S.A.)

coupled with an IBM PC compatible computer using a custom-built 100 gain

amplifier and an analog to digital converter (LABMASTER, Scientific Solutions,

Solon, OH, U.S.A.).

The C02 concentration at the inlet to the leaf chamber was maintained

between 350 and 380 J.'L L -1 by software control using a program written in

ASYST (Asyst Software Technologies Inc., Rochester, NY, U.S.A.). However,

C02 concentrations within the leaf chamber were influenced by the net carbon

fixation rate of the leaf. Leaf chamber C02 concentrations were between 150

and 230 J.'L L-1 (mean.± SD of 28 experiments = 197 .± 32.8 J.'L L-1 C02), as

predicted from measurements of net carbon exchange rate of enclosed leaves,

given a constant flowrate (3.8 L min-1) and constant C02 concentration

(350 J.'L L-1) of air entering the leaf chamber. The implications of low leaf

chamber C02 concentrations will be discussed in the results and discussion

section of this paper.

14C02 was added to the system, as required, through the opening and

closing of a solenoid valve (59-22-900, General Valve Co., Fairfield, NJ, U.S.A.)

which allowed 14C02 to flow into the system from a pressurized source.

Unlabeled C02 could also be used, if unlabeled carbon fixation measurements

were called for. 14C02 of known specific activity was produced by the

57

following method: 1) 14C02 gas of high specific activity was generated in a

sealed vessel by mixing lactic acid with Na214C03 (obtained from Amersham

Co., Arlington Hts., IL, U.S.A.), 2) the 14C02 gas was drawn into an evacuated

lecture bottle, 3) unlabeled C02 was forced into the lecture bottle under

pressure until the amount of cold C02 required to dilute the high specific

activity 14C02 had been added (calculated using the Van der Waals gas

equation at a given volume and temperature), and 4) the precise specific

activity of 14C02 in the lecture bottle was determined by filling the steady state

labelling system (total volume = 17.4 L) with 14C02 to a measured C02

concentration, trapping the C02 in a NaOH trap, and counting aliquots of the

trapping solution.

Whole leaf carbon fixation was calculated from measurements of the

C02 depletion rate within the labelling system, by linear regression of the rate

of C02 depletion (corrected for the internal volume of the labelling system). All

carbon fixation calculations were performed automatically at approximately

1 minute intervals using data stored and analyzed by ASYST software. Leaf

areas were determined by traCing leaves on paper for subsequent

measurement using a ZIDAS digitizer board (Carl Zeiss, Oberkochen, F.R.G.).

Leaf radioactivity was continuously monitored with a Geiger-Muller (GM)

detector (probe model GP200, ratemeter model RLM-2; Wm. B. Johnson,

Research Park, Montville, NJ, U.S.A.) mounted approximately 1 cm from the

abaxial surface of the leaf (counting efficiency of approximately 1.5%).

The relative humidity of air entering the leaf chamber was maintained at

a constant dewpoint of 13 to 18 0 C using a temperature-controlled condenser

and water trap. However, air within the leaf chamber was probably saturated

with water vapor as indicated by the condensation of water on the lower

58

surface of the leaf chamber. The temperature of air within the leaf chamber was

between 23 and 28°C for all experiments, with variation of less than 3°C during

individual experiments. C02 within the steady state system could be removed

using a C02 trap consisting of a fritted glass bubbling tube submerged in a

flask filled with a saturated solution of NaOH, and provided with a cold finger

vapor trap (chilled with crushed ice) to prevent NaOH-laden water vapor from

entering the labelling system.

Experimental procedure. To minimize the effect of plant handling on

the day of an experiment, plants were readied at least 10 h in advance of

experiment initiation. The evening before an experiment, the top leaf of an

individual plant was fastened to a nylon filament support matrix within the open

leaf chamber of the steady state labelling apparatus. The terminal raceme was

removed, and the top axil was trimmed to a single four- or five-seeded pod.

The remaining pod at the top axil was clamped in an inverted position with

loosely fitting holders made from aluminum foil wrapped in Parafilm.

Experiments were conducted in a forced-draft hood, under a 1000 W

metal halide lamp (M1000jCjU Metalarc, GTE Products Corporation,

Manchester, NH, U.S.A.) filtered through 2.5 cm of circulating water in a

Plexiglas bath, providing approximately 1000 j£mol photons m-2 s-1 to the top

leaf. The lamp was turned on between 8:00 and 9:30 A.M., and the leaf was

sealed into the leaf chamber within 9 min of light initiation by surrounding the

petiole with modelling clay, and sealing the two halves of the leaf chamber with

silicon grease. Upon leaf enclosure, steady state labelling commenced, and

was maintained until the end of an experiment, when the plant was dissected

for analysis.

59

In order to test the effect of cutting on phloem transport into seeds, an

incision was made through the dorsal suture of the pod, approximately 20 min

after the initiation of steady state labelling. This incision removed approximately

45% (by fresh weight) of the distal sides of two adjacent seeds, leaving at least

one intact seed on either side of the incision. The surface of the incision was

smeared liberally with silicon stopcock grease, and the entire pod was

surrounded by plastic wrap to prevent drying of the incision. It was observed

that immediately following incision, some of the cut embryos tended to move

slightly above the edge of the cut seed coat.

Analysis of tissues and extracts. All tissues samples (with the

exception of embryos, which were frozen) were placed in test tubes containing

80% (vjv) aqueous ethanol at 50°C, within 15 minutes of plant removal from

the steady state labelling apparatus. Tissues were extracted at least 3 x 3 h

with a total of 6 mL 80% ethanol. Individual leaf punches (1.1 cm in diameter)

from each of the three leaflets of the labeled source leaf were pooled for

extraction. Petiole and peduncle samples were cut to 2 cm lengths, and seed

coats were cut in half to remove the embryo (axis + cotyledons) before

extraction. Ethanol extracts were stored in screw-cap vials in a refrigerator prior

to analysis. An aliquot of each ethanol extract was loaded into tandem Poly­

Prep ion exchange columns (Bio-Rad Laboratories, Richmond, CA, U.S.A.)

conSisting of an upper column of AG 1-X8 resin (formate; 200-400 mesh) and a

lower column of AG 50W-X8 resin (H +; 200-400 mesh). The neutral fraction

was eluted in 80% ethanol and brought to dryness in a 70°C oven. Dried

column eluates were redissolved in 3 x 0.5 mL distilled water, and 0.5 mL of 5%

(w Iv) ZnS04 was added, followed by the addition of 0.5 mL 0.3 N NaOH,

which caused a precipitate to form. The preCipitate was then sedimented by

60

centrifugation at approximately 500 g for 1 min, after which the clarified

supernatant was decanted into a fresh test tube for sugar analysis. The sugar

recovery of this procedure was approximately 90% (±. 3%), based on the

recovery of sucrose standards. The 14C content of aliquots of clarified

supernatant was determined by liquid scintillation counting, and the total

amount of glucose remaining in each sample was determined after sucrose

hydrolysis by invertase using the glucose oxidase/peroxidase, rapid assay

method (Berg meyer and Bernt, 1974). Glucose determination following

inversion quantified the total amount of sucrose + glucose. Following the

passage of extracts through columns, sucrose in the extracts was inverted by

approximately 10% (±. 5%) on a molar basis, as determined by the inversion of

sucrose standards that were passed through columns and assayed for

glucose with and without invertase.

Extracted tissues, ethanol extracts, and neutral column eluates were

counted using a methylcellosolve-based scintillation fluid described by Sun et

al. (1988) in a Beckman Model LS-355 liquid scintillation counter with counting

efficiencies determined by the external standard-channels ratio method. When

ethanol extracts of leaves, petioles or peduncles were counted, 20 JLL of

commercial bleach (Clorox) was added to 0.2 mL of extract in a 7 mL vial at

least 10 min before the addition of scintillation fluid, in order to decolorize the

extract.

Embryos were dried overnight at 60°C, combusted in a Packard Sample

Oxidizer (Tri-Carb B306, Downer's Grove, IL, U.S.A.), and the 14C released

was trapped in Carbosorb II (Packard) for counting using Liquiscint scintillation

fluid (National Diagnostics, Mayville, NJ, U.S.A.) in a Beckman LS-100C

scintillation counter. The 14C recovery of the sample oxidation procedure was

82± 1%.

61

Sugar specific activity calculations. In order to determine the relative

proportions of sucrose, glucose, and fructose in seed coats and leaf disks,

80% ethanol extracts were purified by a non-inverting ion exchange

chromatography procedure similar to that described above, and consisting of a

an upper column of AG 2-X8 resin (formate; 100-200 mesh) and a lower

column of AG 5OW-X8 resin (H +; 200-400 mesh). Neutral column eluates were

collected, dried at 70°C, and rehydrated for spotting on thin layer

chromatography plates (Silica Gel GHL, 20 X 20 cm, 250 microns, Analtech

Inc., Newark, DE, U.S.A.) pretreated with 0.03 M boric acid in 80% (v/v)

aqueous ethanol. Plates were run in two dimensions, using a chloroform,

acetic acid and H20 solvent (6:7: 1; v Iv Iv) for the first dimension, and a solvent

composed of isopropanol and 0.15 M boric acid (4: 1; v Iv) for the second

dimension. Labeled sugars were visualized using autradiography film

(Hyperfilm-BetaMax RPN.9, Amersham Co., Arlington Heights, IL, U.S.A.), and

were identified by reference to standards. Following identification, spots

containing individual sugars were removed from the plates so that the 14C

content of individual sugars could be determined by liquid scintillation counting.

Sucrose, glucose, and fructose comprised greater than 95% of the total 14C in

the neutral column eluates of leaves and seed coats, and glucose and fructose

were present at approximately 4% and 5.5% respectively, of the total sucrose

+ glucose + fructose in the neutral fraction. The relative abundance of

sucrose, glucose and fructose, and the near equivalence of glucose and

fructose in neutral column eluates allowed the calculation of carbon specific

activity in the neutral fraction based on two assumptions: 1) 100% of the 14C

present in the neutral fraction was accounted for by sucrose, glucose, and

fructose, and 2) 12 moles of carbon atoms were present per mole of sucrose

+ glucose assayed.

62

Normalization of tracer data. The specific activity of 14C02 used for

labelling was used to normalize the tracer data of experiments which used

different 14C02 specific activities, so that comparisons between experiments

could be made. This was accomplished by multiplying all 14C and specific

activity measurements within an experiment by a dimensionless quantity

derived by inverting the specific activity of 14C02 used in the experiment, and

multiplying by the units of specific activity (G8q (mol Cr1). Thus, the highest

specific activity that could be reached should be 1.0 G8q mol-1 C.

Analysis of 14C partitioning. Since data collection required plant

harvesting before tissue analysis, multiple plants were labeled for specific times

in order to quantify the progress of steady state labelling. Tissue 14C was

divided into three fractions: 1) neutral fraction (column eluate primarily

containing sugars), 2) non-neutral fraction (calculated as the difference

between neutral 14C and total 14C in extract, and consisting of organic acids

and amino acids), and 3) non-extractable fraction (14C remaining in tissues

following ethanol extraction, and containing primarily starch and structural

carbon). 14C accumulation and seed coat sugar content of cut seeds was

adjusted for reduced seed size by multiplying each cut seed measurement by

the intact seed to cut seed fresh weight ratio determined for the seeds within

each pod. Transport from the leaf was calculated as the difference between net

14C fixation (based on measured net C02 fixation assuming a constant 14C02

specific activity of 1.0 G8q mol-1) and net 14C accumulation in the leaf.

RESULTS AND DISCUSSION

Leaf carbon fixation, accumulation, and export. Photosynthesis was

maintained at fairly constant rates throughout the labelling period, with slight

declines «20%) observable after 8 h in some experiments (Fig. 2.2). The net

63

Figure 2.2 Net carbon exchange rate and 14C accumulation by a single leaf.

Measurement of net carbon exchange rate and 14C accumulation (using GM

detector) are described in Materials and Methods. Zero time refers to the

beginning of steady state labelling. Single representative experiment.

N

<l

<l

64

<I

<I

<l

<I <I

<I

<I

a

co ~ N a (L_s Z __ UJ Z08lowrl)

8lDJ 85uD40X8 uOqJDO l8N "

N

co ~

rn L ::l 0

..c: '-..J

Q)

E .-f-

~

a

65

carbon exchange rates observed in these experiments (mean.±. SD of 28

experiments = 5.14.±. 0.985 ",mol C02 m-2 s-1) were similar to rates reported

for Phaseolus vulgaris under the same C02 concentrations rJassey and

Sharkey, 1989). Continuous measurement of 14C accumulation in leaves with

a GM tube demonstrated nearly constant rates of radioactivity increase during

the labelling period (Fig. 2.2). Slight decreases in the slope of the radioactivity

vs. time curve « 15%) were sometimes observed after approximately 5 h of

labelling. The observation of a relatively continuous increase in 14C

accumulation within the leaf (Fig. 2.2) may be attributed to a combination of

increasing specific activity of leaf carbohydrate pools (presumably to a

maximum equivalent to the specific activity of the 14C02 used for labelling),

and the storage of 14C in the leaf as starch.

Partitioning of 14C within the leaf is illustrated in Figure 2.3. The rate of

net 14C fixation by photosynthesis in 14C02 with a specific activity of

1.0 G8q mol-1 was 19.3 M8q 14C m-2 h-1, as determined by linear regression

of data from Figure 2.3A (r2 = 0.988). The rate of net 14C accumulation within

the leaf was 8.0 M8q 14C m-2 h-1 also determined by linear regression of data

in Figure 2.3A (r2 = 0.989). The high r2 values determined by linear regression

demonstrate that both net 14C fixation and net 14C accumulation by the leaf

were linear over time, and provide evidence for relatively constant rates of net

carbon fixation and accumulation during the labelling period. The calculated

rate of 14C transport from the leaf also appeared to be constant over time at a

value of 11.3 M8q 14C m-2 h-1 (r2 = 0.976, Fig. 2.3A). When expressed as a

percentage of net 14C fixation, 14C transport from the leaf was 53.7 .±. 2.1 % of

the total 14C fixed (mean.±. SE of 28 measurements, see Fig. 2.38). This value

is comparable to published data for the allocation of carbon to export during

66

Figure 2.3 14C Partitioning in source leaves over time. (A) time course of net

14C fixation, 14C transport from the leaf, and 14C accumulation in the leaf. (8)

14C transport from the leaf as a percent of the total 14C fixed by the leaf. (C)

partitioning of total 14C fixed by the leaf between the neutral fraction (sugars),

non-neutral fraction (organic acids + amino acids), and the non-extractable

fraction (starch + structural). Data for (C) expressed as percentage of total

14C fixed by the leaf. Points represent the mean.±. SE of three experiments,

except for 13 h points which are the mean of.±. SE of 16 experiments. Tracer

data were normalized to a 14C02 specific activity of 1.0 GBq mol-1 (see

Materials and Methods). Zero time refers to the beginning of steady state

labelling.

67

250 A-Leaf 14C partitioning N I 200 E cr 150 m :2 100 u ~

..- 50

• Net fixed o Transported • Remaining in leaf

~ 8- 14C transport from leaf T

'-" 60 T / 0- _ ~ -0- - - -0/ 1 ..... ~D .: 50 0- - 1 .1

u :! 30

o -+-' o

-+-' 20 4-o

-+-' C Q) 1 0 U L ClJ

0..

c- 1 4C in leaf • Neutral o Non-Neutral

• . • • Non-'-Extractable - 1 _______

Q---- ~--. I o A~~-~O . .------ ~

o~~~~~--~~~~~--~~~~~

o 4 8 12 Time (hours)

68

the photoperiod, which range from 40 to 80% for an assortment of species,

and shows considerable variability even within a single species (between 47

and 72% for Glycine~; Gordon, 1986). Overall, the rates of 14C fixation,

export, and accumulation within the leaf observed in these experiments were

relatively constant over time, and agree with published values. However, it

must be noted that net carbon fixation in the labeled leaf was limited by low

C02 concentrations.

Leaf 14C and sugar partitioning. The greatest proportion of total fixed

14C remaining in the leaf was present in the neutral fraction at all times during

the labelling period, which indicates that most of the fixed 14C was present in

sugars (Fig. 2.3C). An increasing proportion of fixed 14C was present in the

non-extractable fraction (Fig. 2.3C), which was most likely due to the

accumulation of 14C in starch (Grange, 1985). The time course of 14C

partitioning in the non-neutral fraction paralleled the neutral fraction, although

as a total percentage of fixed 14C, the non-neutral fraction was significantly

lower (Fig. 2.3C). The non-neutral fraction, which represents primarily amino

acids and organic acids, was a greater proportion of the total extractable 14C

in leaves (44% of extractable 14C), as opposed to all other tissues measured in

these experiments, which did not differ significantly from each other in the

proportion of 14C in the non-neutral fraction (approximately 28% of extractable

14C). The presence of a relatively large amount of fixed 14C in the non-neutral

fraction of leaf tissues was probably due to the large pool of organic and amino

acids which serve as the intermediates and by-products of photosynthetic

metabolism in leaves (Kouchi and Yoneyama, 1984b).

Sugar content (sucrose + glucose) of the leaf was variable over time,

but no statistically significant differences between times were observed using

one way analysis of variance with 95% confidence. The lack of significant

69

changes in leaf sugar content over time supports the observations of Fondy et

al. (1989) which demonstrate that there is no significant sucrose accumulation

in Phaseolus vulgaris leaves beyond 3 hours after the initiation of a square­

wave light regime. The average sucrose + glucose content of the leaf was

4.67 mmol m-2 (28 measurements, SE = 0.39) which is comparable to the

values reported by Fondy et al. (1989).

Changes in leaf sugar specific activity over time. Measurements of

the leaf sugar specific activity over time (Fig. 2.4A) appeared to follow

saturation kinetics as reported previously for bean, sugar beet and squash

(Geiger, 1980). However, the time required to reach isotopic saturation (no

further change in specific activity) in these experiments were longer than the

90 min time presented by Geiger (1980). The leaf sugar pool appeared to be

saturated with 14C after 5 h of labelling (Fig. 2.4A). The relatively slow

equilibration of leaf sugar with tracer observed here may have been due to a

significant unlabeled sugar source present in the leaf, or may indicate that the

14C02 specific activity was reduced at the beginning of experiments due to

leaf respiration or to the initial unlabeled C02 present in the leaf chamber. The

fact that the leaf sugar pool saturated at a specific activity value approximately

equal to 75% of the supplied 14C02 (1.0 GBq mol-1) is not highly unusual, in

that this property has been demonstrated in similar experiments in the literature

(85% for sugar beet leaf sucrose: Geiger, 1980; 62% for soybean leaf sucrose:

Fisher, 1970b). Isotopic discrimination against 14C is typically offered as the

reason for low leaf sugar specific activities (relative to the supplied 14C02)

(Geiger, 1980). However, it is possible that other factors, such as the presence

of a pool of slowly labeled sugar (Fisher 1970b, Geiger et aI., 1983), may also

be relevant to this observation.

70

Figure 2.4 Tissue sugar (sucrose + glucose) specific activity versus time.

Symbols for A and B: leaf (e), petiole segment near leaf (.6.), petiole segment

near pod (v), and peduncle of pod (0); C and D: intact seed coat (_) and cut

seed coat (0). A and C illustrate the sugar specific activity of the tissues, while

Band D display the ratio of tissue specific activity to leaf specific activity for

each tissue. Points represent the mean ± SE of three experiments, except for

13 h points which are the mean of ± SE of 16 experiments. Specific activity

data were normalized to a 14C02 specific activity of 1.0 GBq mol-1 (see

Materials and Methods). Zero time refers to the beginning of steady state

labelling.

71

A-Leaf & stem S.A. B-Stem/leaf :::' 0.8 •

T • I • ______ •

E 0.6 T ,..,..,1 T

0" /' •. ::~

S.A. ratio I .... ~ fl. .' <> T •

A. .v .. ' ,

OJ 0.4 1 .... T '.

(!) fl.' . "tAo A'·· '-' . "'yi' 8" Z'0.2 ... :.~::' ...

~ /t:'" « C-Seed coat S.A. (J

't= 0.5 .-

I •

fl. :.<>. 8:.. ". . .. : 1 '. v ... :.0'

'9'"

TO-Seed coat/leaf • S.A. ratio

4 8 12 4 8 Time (hours)

12

1 .0 ~

0.8 > ofJ

0.6 ~ (J

0.4 !+-(J

0.2 ~ (f)

~

c Q)

.....J 0.8 ~

o

0.6 5 .-ofJ (J

0.4 ~ l.L.

0.2

72

Calculation of the sugar export rate from the leaf. The rate of sugar

export from the leaf could be predicted from the rate of 14C export, because

the specific activity of leaf sugar remained at an approximately constant value

of 0.65 GBq mol-1 following 2.5 h of labelling (see Fig. 2.4A). Under the

assumption that 100% of the 14C exported from the leaf was in sucrose, and

using the leaf 14C export rate of 12.6 MBq m-2 h-1 calculated by linear

regression of the transport data in Figure 2.3A (omitting the 2.5 h pOint

because a steady state had not been reached; regression r2 = 0.954), the

calculated rate of sugar export was 1.6 mmol m-2 h-1. This value is in good

agreement with previously reported values for the rate of sucrose export from

leaves of Phaseolus vulgaris that were also calculated assuming 100% sucrose

export, and using steady state labelling measurements:

1.6 mmol sucrose m-2 h-1 for two week old seedlings (Fondy and Geiger,

1980), and 1.9 mmol sucrose m-2 h_1 for plants with developing pods (Fondy

and Geiger, 1983). The assumption that 100% of the 14C exported from the

leaf is represented by sucrose is an overestimate. However, greater than 90%

of the label in the neutral fraction (sugars) of leaves was present in sucrose as

determined by thin-layer chromatography, and as published for soybean

leaves and petioles (Fisher, 1970a). In conclusion, the rate of leaf sugar export

in these experiments could be calculated within a reasonable margin of error

using the observed leaf 14C export rate.

Specific activity of sugars in the path and seed coat tissues. The

specific activity of sugar in tissues along the path of photosynthate transport

from the leaf to the seeds is illustrated in Figures 2.4A and B. Tissues of the

transport pathway that were in closer proximity with the source of

photosynthate (Le. the petiole sample near the leaf vs. the petiole sample near

73

the stem, vs. the peduncle sample) had higher sugar specific activities than

those farther from the source at all times after the onset of steady state

labelling (Fig. 2.4A). In addition, the pathway tissues closest to the source leaf

had the highest relative sugar specific activities in comparison with those of the

source leaf (Fig. 2.48). The fact that the specific activities of sugar in the path

tissues nearer to the source were closer to the steady state specific activity of

the leaf would be expected due to the shorter time for movement of 14C from

the leaf to the tissues nearest to the source, and could also be caused by

increasing exchange with unlabeled sugar along the path from source to sink.

It is also apparent from Figures 2.4A and 8 that the path tissues reached

relatively high specific activities early during the steady state labelling period, in

parallel with those of the source leaf. This would be expected for tissues in

which sugar is present primarily in a pool that equilibrates rapidly with the leaf

sugar pool, such as the phloem transport stream. Evidence supporting the

hypothesis that the greater part of petiole sugar is contained within the phloem

transport pool is provided by Fisher (1970a), who used pulse-labelling to

demonstrate that the specific activity of sugar in the petioles of kidney bean

and soybean rapidly reached equilibrium with that of the leaf, and calculated

that approximately two thirds of the sugar in the petiole was present in the

phloem transport stream.

During the course of steady state labelling, the seed coat and path

sugar pools never reached complete isotopic equilibrium with the leaf sugar

pool, as demonstrated by the observation that the fraction of leaf sugar specific

activity attained by path and seed coat tissues was always less than 0.8 (Figs

2.48 and D). This may have been due to isotopic discrimination against 14C by

74

some process involved in the entry of sugar into the path or seed coat sugar

pools, but, more likely, the lack of complete isotopic equilibration with the leaf is

due to the presence of sugar pools in these tissues which exchange very

slowly with the sugar of the phloem transport pool.

The specific activity of sugar in the path tissues was significantly greater

than that of the seed coat following 2.5 h of steady state labelling. However, the

gap between the path and seed coat sugar specific activities narrowed over

time (Fig. 2.4). After 5 h of steady state labelling, the specific activity of the seed

coat sugar pool became quantitatively similar to those of the path tissues, and

remained lower over time, to approximately the same extent that the specific

activities of path tissues were lowered by their relative distances from the

source leaf (Fig. 2.4). In summary, after 5 h of steady state labelling, the

specific activities of the path and seed coat tissues appeared to differ primarily

due to their unequal proximity to the source. However, the specific activity of

seed coats was relatively low in comparison to path tissues during the initial

2.5 hours of steady state labelling. One possible explanation for the low seed

coat sugar specific activity at the beginning of steady state labelling is that the

sugar pool in the seed coat has a slow turnover time caused by a large pool

size relative to the flux through the seed coat. It is likely that, in comparison with

path tissues, the sugar pool of the seed coat is large relative to the flux through

the tissue, because the seed coat is a sink tissue in which the unloading of

sugar from the phloem transport stream may be significant (Patrick and

McDonald, 1980; Thorne, 1980; Gifford and Thorne, 1986). If the sugar

imported by the phloem must exchange with the bulk of the sugar in the seed

coat before it is transferred from the seed coat to the embryo (the symplastic

unloading route hypothesis), then this would explain the relatively slow initial

equilibration of the seed coat sugar pool with tracer.

75

Carbon partitioning within intact seeds. The time course of 14C

import into intact seeds is illustrated in Figures 2.5A and B. During the first

2.5 hours of labelling, the 14C content of the seed coat was greater than in the

embryo, an observation that appears to be similar to the relatively slow radial

transfer of tracer carbon between the seed coat and embryo that had been

described previously in pulse-labelling experiments (Thorne, 1980; Patrick and

McDonald, 1980). Slow radial transfer of tracer carbon through the seed coat

may be caused by the lateral movement of tracer throughout the entire seed

coat before transfer to the embryo as described above for the symplastic

unloading route hypothesis, and also by Thorne (1980) and Patrick and

McDonald (1980).

The accumulation of 14C within embryos was greater than accumulation

within seed coats after 5 hours of labelling, and there was an increase over

time in the proportion of 14C in the embryos versus the seed coat (Fig 2.5A

and B). Given that the embryos are an ultimate sink for carbon imported from

the leaf, the continued increase in embryo 14C content would be expected. In

contrast, the seed coat is both a sink tissue and a tissue through which

photosynthates pass on their way into the embryo. In keeping with the role of

the seed coat in the transfer of sugars from the phloem to the embryo, the

majority of seed coat 14C was present in the neutral (sugar) fraction (Fig.

2.5B). The increase in total seed coat 14C content over time was primarily due

to an increase in sugar specific activity (Fig. 2.4C), and only a small and

decreasing proportion of 14C accumulated in the non-neutral (amino and

organic acid) and non-extractable (starch + structural) fractions of the total

seed coat 14C (Fig. 2.5B). The fact that very little 14C accumulated in the non-

76

Figure 2.5 Partitioning of 14C in intact and cut seeds over time. A and C, total

14C in seed (0), embryo (e), and seed coat (A.); 8 and 0, percent oftotal 14C

in seed for embryos (e), seed coat neutral fraction (A.), seed coat non-neutral

fraction (.6.), and seed coat non-extractable fraction (\7). Data presented on a

per seed basis, as the mean of two seeds per pod for the cut seed treatment

(C,D), and mean of two to four seeds per pod for the intact seed treatment

(A, 8). Points represent the mean ± SE of three experiments, except for 13 h

points which are the mean of ± SE of 16 experiments. Tracer data were

normalized to a 14C02 specific activity of 1.0 G8q mol-1 (see Materials and

Methods). Zero time refers to the beginning of steady state labelling.

300 A-Intact, Total 14C

I 200 -0 Q) Q) U')

0"100 CD ~ ......., Q) ::J U') U')

.c 100 u ~

.s 50 o ~

c-cut, Total 14C

IJ o

4 8

77

Q 8-lntact, Percent 14C ~'" 'P __ •

:x·-' . ~-/J. '-~--~

·--A ---A-A-A

O-Cut, Percent 14C

~,­~,

~-~ .........

. .,......,... A/~-A_A •

12 4 8 12 Time (hours)

40 ~ Q) U')

20·c

u ~

o 80 -0

-+-' .....

60 0 -+-' C Q)

40 0 '­Q)

a.. 20

78

extractable fraction (Fig. 2.58) indicates that the seed coats accumulated very

little or no starch during the labelling period, and suggests that seed coat

starch is probably not a significant source of carbon for later mobilization

(Fader and Koller, 1985).

Calculation of the rate of sugar import into intact seeds. The

hypothesis that the flux of carbon from the labeled leaf to the seeds at the

subtending node is constant and is the only source of carbon for seed import

is central to the determination of the rate of sugar import into intact seeds that

will be presented in the next paragraph. Support for the hypothesis that the

carbon flux from the labeled leaf remained constant during the labelling period

was presented in the discussion of leaf sugar export rate. Pulse-labelling

experiments have indicated that the primary source of photosynthate for the

pods at the top axil is the top leaf in Phaseolus vulgaris (Wien et aI., 1976;

Lucas et al., 1976), and the production of photosynthate by the pods

themselves could not provide a significant source of carbon for import into

developing Phaseolus vulgaris seeds (Oliker et aI., 1978). During the course of

steady state labelling, the labeled leaf received much greater illumination than

the remaining leaves of the plant, and was probably the major source of

photosynthate for the entire plant. Also, the removal of pods on the terminal

raceme and top axil reduced the sink load of the labeled leaf, and would thus

be expected to reduce sink demand from the unlabeled leaves of the plant.

Although there may have been some flux of carbon from the unlabeled leaves

to the seeds of the pod at the top axil, the conditions of these steady state I

labelling experiments would appear to minimize this flux in comparison with the

flux of carbon from the labeled leaf. The observation that the specific activity of

79

seed coat sugar reached a relatively high proportion of the leaf specific activity

(approximately 75% after 13 h of labelling, see Fig. 2.40) is further evidence

that at least the majority of carbon flux into the seeds was provided by export

from the labeled leaf.

Assuming that a constant flux of photosynthate was provided by the

labeled leaf, the rate of sugar transport into the embryo of intact seeds could

be calculated based on the sugar accumulation within the embryos over time

as determined from the embryo 14C content and the specific activity of sugar

transported to the embryo. In order to determine the specific activity of sugars

transported to the embryo, the pathway of photosynthate movement from the

seed coat phloem to the apoplast of the embryo must be taken into account. If

sugars are unloaded directly out of the seed coat phloem, with little or no

mixing with the symplastic sugar pool of the seed coat (the apoplastic

unloading route hypothesis), then the specific activity of sugar accumulated by

the embryo would be expected to be similar to that of the leaf sugar pool (i.e.

the phloem transport stream). If phloem sugar must pass through and mix with

the entire symplast of the seed coat before uptake by the embryo (the

symplastic unloading route hypothesis), then the specific activity of sugar taken

up by the embryo would be similar to the specific activity of the seed coat

sugar pool. It is important to note that if a significant quantity of sugar were

present in a pool that did not exchange rapidly with the seed coat symplast

(such as a vacuolar pool), then the conditions of the apoplastic unloading route

hypothesis would be met, in that the specific activity of sugars transported to

the embryo would be similar to the specific activity of sugar in the leaf phloem

transport pool.

80

Through application of the symplastic unloading route hypothesis, the

rate of sugar import to the embryo was determined to be 2.53l'mol sucrose

seed-1 h-1 (r2 = 0.8027), as calculated from the specific activity of seed coat

sugar (see Fig. 2.4C) and the 14C content of the embryo. The rate of sugar

import to the embryo determined by similar application of the apoplastic

unloading route hypothesis was 2.321'mol sucrose seed-1 h-1 (r2 = 0.9965),

using the specific activity of the leaf sugar pool (see Fig. 2.4A) and the 14C

content of the embryo. The rates calculated above were based on the

assumption that 100% of the 14C taken up by the embryo consisted of

sucrose, while Patrick and McDonald (1980) have reported that, following a

pulse label, approximately 80% of the 14C photosynthate present in the

cotyledonary apoplast was labeled sucrose. Allowing that 80% of 14C imported

by the embryo was labeled sucrose, the rates of sucrose accumulation by

embryos were 1.86 and 2.021'mol sucrose seed-1 h-1 respectively, for the

apoplastic and symplastic unloading hypotheses.

The rate of sucrose accumulation in the embryos of a small seeded

variety of Phaseolus vulgaris (250 mg seed-1) was 2.3 I'mol sucrose

seed-1 h-1, as determined using in vitro sucrose uptake measurements

(Patrick, 1981), and other studies have reported rates of embryo dry weight

accumulation between 0.71 and 1.33 mg seed-1 h-1 during the period of rapid

seed fill for Phaseolus vulgaris seeds with final dry weights between 250 and

400 mg (ct. Walbot et aI., 1972; Hsu, 1979; Patrick, 1981). Under the

assumption that approximately 80% of final seed dry weight enters the seed as

sucrose (Patrick and McDonald, 1980), the rate of sugar import into the

embryo was between 1.7 and 3.1 I'mol sucrose seed-1 h-1 as calculated from

the dry weight accumulation rates given above. The final dry weight of seeds in

81

this study was 550 mg, although the period of seed fill was no longer than for

the cultivars reported above (approximately 36 days). The fact that the cultivar

used in these experiments had a greater seed dry weight than those reported

for other cultivars, without an increase in the seed fill period, argues that the in

vivo rate of sugar accumulation may have been greater for the seeds in the

current study. It is apparent that the sugar import rates calculated employing

either the symplastic route hypothesis (2.02 "mol sucrose seed-1 h-1) or the

apoplastic route hypothesis (1.86 "mol sucrose seed-1 h-1) were within the

lower range of the rates calculated above for cultivars of lower seed dry weight.

The expectation of a higher rate of sugar import to the embryo in these

experiments relative to those calculated from earlier studies indicates that there

may be some factor which lowered the calculated rate of photosynthate import

into the intact seeds of these experiments. In addition, no statistically significant

difference was observed between the predicted sugar import rates from the

two unloading route hypotheses, due to the small size difference and the

inherent variability of the calculated rates. The symplastic unloading route

hypothesis appeared to produce a more accurate characterization of sugar

import by the embryo because the linear regression derived using this

hypothesis could be extrapolated through the origin (zero sugar accumulation

at zero time), while the regression line derived using the apoplastic unloading

route hypothesis did not pass through the origin, as determined using a t-test

of significance for a straight line fit through the origin, at the 92% confidence

level (Snedecor and Cochran, 1980). Even though the application of the

symplastic unloading route hypothesis appeared to produce a more accurate

description of sugar accumulation within the embryo, no strong conclusions as

to the primacy of one unloading route over the other may be made from the

observations described above.

82

The effect of cutting on photosynthate transport into seeds. The

time course of 14C accumulation within surgically modified seeds is presented

in Figures 2.5C and D. In order to compare the intact and cut seed treatments,

the amounts of 14C and sugar within the various cut seed fractions were

adjusted using a correction factor based on the fresh weight of the cut versus

intact seeds within each pod (the typical seed size correction factor was

approximately 1.6, based on an average of 62% seed fresh weight remaining

after cutting, see Materials and Methods). The pattern of 14C accumulation

within the seed coat and embryo of cut seeds was the opposite of that

observed for intact seeds, as the 14C content of the embryo was lower than

the 14C content of the seed coat at all times during the labelling period (Fig.

2.5C). There was an observable increase over time in the proportion of 14C

accumulated within the embryo in comparison with the seed coat. However,

the 14C content of the embryo never surpassed that of the seed coat, as it did

in intact seeds (Fig 2.5). The lowered embryo 14C content observed in the cut

seeds indicates that there may have been an inhibition of 14C movement into

the embryo of cut seeds as will be discussed further below.

The statistical significance of comparisons between intact and cut seeds

with respect to the sucrose content of the seed coats, the specific activity of

sugar in the seed coat, and the 14C content of various fractions are presented

in Table 2.1. There were statistically significant differences between time

periods regarding seed coat sugar specific activity and the 14C content of all

seed fractions, as demonstrated by the small p-values calculated for the time

period portion of the analysis of variance in Table 2.1. The calculation used to

determine the statistical significance of differences between time periods

combined the intact and cut seed data at each time period, and thus the

TABLE 2.1 Statistical significance of differences between intact and cut seeds with respect to 14C and sugar

(sucrose + glucose) accumulation. Experimental design consisted of 12 experiments divided into four time periods

(2.5,5,8, and 10 h) with three replications per time period. Each experiment contained two cut seeds and two to

four intact seeds which were pooled to produce a cut and an intact treatment for each experiment. The statistical

significance of differences between cut and intact seeds, between time periods, and for the interaction between

time periods and differences between cut vs. intact seeds, is presented as the F-test p-values determined using

two-way analysis of variance (two-way crossed model, Snedecor and Cochran, 1980). The smaller the p-value, the

greater the effect of the treatment (i.e. the larger the component of variability attributable to the treatment). With the

exception of seed coat sugar content, the data used for this table were presented in Figures 2.4 and 2.5. Tracer

data were normalized to a 14C02 specific activity of 1.0 GBq mol-1 (see Materials and Methods).

Component Seed coat Seed coat of sugar sugar

variability content S.A.

Cut vs. Intact 0.000** 0.091/

Time Period 0.184 0.003**

Cutting x Time 0.625 0.639 Interaction

/ Significant at the 90% confidence level.

* Significant at the 95% confidence level.

** Significant at the 99% confidence level.

14C in seed coat fractions 14Cin Total 14C

Neutral Non-Neut. Non-Extract. Total Embryo in Seed

0.714 0.554 0.901 0.617 0.032* 0.142

0.010** 0.004** 0.001** 0.006** 0.015* 0.010**

0.955 0.989 0.988 0.986 0.322 0.707

~

84

differences between time periods analyzed using this method are equivalent to

the combination of the intact and cut seed data for each time point presented

in Figure 2.5. Statistical analysis with respect to the differences between time

periods only highlighted time trends that could be readily observed in Figure

2.5. However, the analysis also indicated that there was no statistically

significant difference in the sugar content of seed coats over time, as indicated

by the relatively large p-value calculated for the time period portion of the seed

coat sugar content analysis of variance in Table 2.1.

In all of the comparisons in Table 2.1, there was no statistically

significant effect of time on the relative differences between cut and intact

seeds, as seen in the large p-values calculated for the cutting x time interaction

portion of Table 2.1. This indicates that the differences between intact and cut

seeds were constant over time and could thus be analyzed by pooling data

across time periods to produce a cut and an intact treatment for each seed

fraction. The statistical significance of the differences between cut and intact

seeds pooled across time periods are presented in the cut vs. intact portion of

Table 2.1, and the pooled averages for cut and intact seed treatments are

presented in Table 2.2.

The observation that the unlabeled sugar content of cut seed coats was

greater than that of intact seed coats (Tables 2.1 and 2.2) may be at least

partially explained by the uneven distribution of sugar in seed coats. The basal

portion of freshly cut seed coats contained more sugar than the distal half (the

basal 62% of the seed by fresh weight contained 70% of the total seed coat

sugar), and thus the application of a correction factor based on the removal of

a given proportion of the cut seed fresh weight, when applied to the sugar

remaining in the basal portion of the seed coat tended to overestimate the total

85

TABLE 2.2 The effect of cutting on 14C and sugar (sucrose + glucose)

accumulation in seeds. Data are pooled averages for intact and cut seed

treatments across time periods, using the data set analyzed in Table 2.1. The

relative size of differences between cut and intact seeds can be observed here.

The statistical significance of these differences is presented in Table 2.1. Tracer

data were normalized to a 14C02 specific activity of 1.0 G8q mol-1 (see

Materials and Methods).

Seed coat

Sugar Sugar Total 14Cin Total 14C

Seed content S.A. 14C embryo in seed

Treatment (JLmol) (G8q mol-1) (k8q) (k8q) (k8q)

Cut 18.86 0.116 40.95 16.58 53.65

Intact 13.82 0.177 36.25 47.24 87.35

86

seed coat sugar content of cut seeds. In the time course experiments, the

basal 55% of the seed (by fresh weight) contained an average of 73% of the

intact seed coat sugar, and thus it appeared that more sugar was present in

the cut seed coats than could be accounted for by the application of the fresh

weight correction factor alone. Other explanations for the calculated excess of

sugar in cut seed coats include the accumulation of sugar in the cut seed coats

as a response to wounding, or due to a blockage in sugar transport to the

embryo, or due to a decrease in sugar uptake by the embryo. The lack of

statistically significant changes in cut seed coat sugar content over time

indicates that if wounding or other responses to cutting the seed were

responsible for the increase in cut seed coat sugar content, then these

responses must have occurred during the first 2.5 hours of experiments. The

increased sugar content observed for cut seed coats was probably due to a

combination of sugar content overestimation by the seed size correction factor

and the accumulation of sugar in cut seed coats due to wounding or to an

inhibition of transport to, or uptake by, the embryo.

The specific activity of sugar appeared to be lower in cut seed coats

than in intact seed coats, although differences were not significant at the 95%

confidence level (Tables 2.1 and 2.2). There was no effect of the seed size

correction factor on the sugar specific activity measurement, and thus the

observed difference between intact and cut seed coats could not be due to

seed size correction. The lower sugar specific activity of cut seed coats may be

explained by either a lower 14C level in the seed coat sugar pool, or by the

accumulation of unlabeled sugar in the seed coat, or by both lowered 14C and

increased sugar content. A lowered 14C content in the cut seed coat sugar

pool could be caused by an inhibition of 14C import into the cut seed coats,

87

but this effect was not observed (Table 2.2). The sugar content of cut seeds

was enhanced in comparison with intact seeds as described above, and this

could provide an explanation for the low seed coat sugar specific activity.

Another explanation could be that some of the seed coat was damaged by

cutting, such that 14C-sugar movement from the phloem to the damaged

portion of the seed coat was restricted. If this were the case, the damaged

portion of the seed coat could contain sugar at a low specific activity.

There were no statistically significant differences between cut and intact

seeds with respect to the 14C content of any of the various seed coat fractions.

However, there was a large difference between the cut and intact embryos with

respect to their 14C content (Tables 2.1 and 2.2). The lack of statistically

significant differences between the 14C contents of cut versus intact seed

coats provides evidence for the hypothesis that 14C movement to the seed

coats was not dramatically inhibited by cutting the seed. The fact that the 14C

content of cut embryos was much lower than that of the intact embryos (Table

2.2) could be explained by the inhibition of 14C movement into the embryos of

cut seeds. The inhibition of 14C movement into cut embryos but not into cut

seed coats suggests that in response to cutting, the movement of

photosynthate between the seed coat and the embryo was inhibited, while

import into the seed coat itself was not greatly inhibited. An alternative

explanation for the relatively low 14C content observed in cut embryos is that

the amount of tissue removed by cutting was inherently unequal in cut

embryos and cut seed coats, and application of the same seed size correction

factor to both tissues distorted the relative differences between embryos and

seed coats. Although it is not improbable that unequal adjustment of cut seed

data could account for some of the difference between cut embryos and seed

88

coats, it is doubtful that this factor could be completely responsible for the large

difference in response to cutting of seed coats and embryos because the

relative amounts of tissue removed by cutting appeared to be comparable for

embryos and seed coats. Another alternative to the theory that the reduced

14C content of the cut embryos was caused by the inhibition of transport

between the seed coat and the embryo, is the possibility that the cutting

treatment specifically inhibited the uptake of photosynthate by the embryo.

There is no apparent reason why cutting the seed would inhibit photosynthate

uptake to a greater extent in the embryo than it would in seed coats, especially

when considering the extent of phloem damage that would be expected when

. cutting into the seed coat. Thus, it is doubtful that a large proportion of the

decrease in 14C accumulation by cut embryos was due to a specific inhibition

of photosynthate uptake by the embryos themselves. The best supported

explanation for the decrease in 14C accumulation by cut embryos appears to

be the inhibition of photosynthate transfer between the seed coat and embryo

of cut seeds. Further evidence for the inhibition of photosynthate transfer

between cut seed coats and embryos was provided above by the observation

that the sugar content of cut seed coats was greater than in intact seed coats

and that the seed coat sugar specific activity was lower in cut seed coats

[fable 2.2).

One explanation for the inhibition of photosynthate transfer between the

seed coat and embryo of cut seeds is that because the embryos of some cut

seeds moved slightly with respect to the seed coats after cutting in some seeds

(see Materials and Methods), the loss of intimate contact between seed coats

and embryos might be responsible for the inhibition of photosynthate transfer.

89

Another hypothesis is that the cutting treatment had a general inhibitory effect

on photosynthate unloading from the seed coat. These two hypotheses cannot

be distinguished by the experiments presented here.

CONCLUSIONS

The use of steady state labelling in these experiments produced

relatively stable fluxes of labeled carbon which were useful for the analysis of

photosynthate transport into the developing seeds of Phaseolus vulgaris.

Although no definitive conclusion could be reached as to the pathway of

photosynthate movement from the phloem to the apoplast of the seed coat,

the behavior of tracer carbon within the seed coat produced somewhat

different predictions for the apoplastic and symplastic unloading route

hypotheses. The observations presented here were best explained by the

symplastic unloading route hypothesis. In addition, these experiments provided

evidence for the disruption of photosynthate transport by the surgical excision

of the distal halves of seeds, as practiced during the preparation of "empty"

seed coats. The use of quantitative methods for the analysis of tracer

movement within developing seeds may provide additional tools for the study

of the mechanisms regulating phloem unloading within developing legume

seed coats.

CHAPTER 3

Turgor-Sensitive Photosynthate Unloading from Perfused Seed

Coats

INTRODUCTION

The empty seed coat technique permits the analysis of photosynthates

released from maternal seed tissues following the surgical removal of the

embryo and the addition of a trapping solution or agar (Thorne and Rainbird,

1983; Wolswinkel and Amerlaan, 1983; Patrick, 1983; Porter et aI., 1985).

Sucrose is quantitatively the most important material eluted from perfused seed

coats, and has been the focus of most experiments (Thorne and Rainbird,

1983; Wolswinkel and Amerlaan, 1983; Patrick, 1983; Thorne, 1985;

Wolswinkel, 1988). Many studies involving the empty seed coat technique have

endeavored to resolve the mechanisms governing photosynthate unloading in

developing legume seed coats, and a great deal of information about the

properties of these mechanisms has been obtained (for reviews see: Thorne,

1985; Murray, 1987; Wolswinkel, 1985b, 1988).

The efflux of pulse-labeled photosynthates and unlabeled sugars (and

amino acids) from attached seed coats was enhanced by perfusion solutions

containing high concentrations of an osmoticum (Wolswinkel and Ammerlaan,

1984, 1985a, 1986; Wolswinkel et al. 1986; Minchin and McNaughton, 1986;

Ellis and Spanswick; 1987; Grusak and Minchin, 1988). Contrary to these

observations, low osmoticum concentrations in the perfusion solution

increased the efflux of photosynthates from excised seed coats (Patrick, 1983,

1984; Patrick et al. 1986; Wolswinkel and Ammerlaan, 1986; Wolswinkel et aI.,

1986; Grusak and Minchin, 1988), and attached seed coats (Minchin and

90

91

McNaughton, 1986). The conflicting results obtained with excised versus

attached seed coats have been explained by the hypothesis that the net efflux

of photosynthates from attached seed coats is dominated by phloem import,

while the net efflux of photosynthates from excised seed coats is controlled by

a mechanism acting on the efflux itself (Wolswinkel et aI., 1986; Grusak and

Minchin, 1988).

The mass flow hypothesis of phloem transport predicts that lowering the

turgor pressure of sieve elements within sink tissues will enhance phloem

import into sinks (Munch, 1930; Lang, 1983). Thus, the enhanced efflux of

photosynthates associated with high osmoticum concentrations was explained

by an increase in phloem import due to low cell turgor in attached seed coats

(Wolswinkel and Ammerlaan, 1984, 1985a, 1986; Wolswinkel et aI., 1986;

Minchin and McNaughton, 1986; Ellis and Spanswick, 1987; Grusak and

Minchin, 1988). Osmoticum concentration changes did not modify membrane

permeability as determined from 14C-mannitol uptake experiments (Patrick et

aI., 1986). Therefore, the increase in photosynthate efflux at low osmoticum

concentrations could be explained by either an increase in efflux through a

turgor-sensitive carrier (Patrick, 1984; Patrick et aI., 1986), or by a decrease in

uptake by a turgor-sensitive apoplastic retrieval mechanism (Wolswinkel and

Ammerlaan, 1986; Wolswinkel et al. 1986). Photosynthate efflux control by a

turgor-sensitive mechanism (the turgor homeostat hypothesis) has received

substantial support from 14C-photosynthate efflux studies (Patrick, 1984;

Patrick et aI., 1986). Evidence for turgor-sensitive sugar retrieval was obtained

from labeled and unlabeled sugar efflux measurements, but this evidence is

circumstantial as turgor-sensitive sugar uptake was not measured (Wolswinkel

92

and Ammerlaan, 1986; Wolswinkel et aI., 1986). This study will attempt to

distinguish between the apoplastic retrieval and the turgor homeostat

hypotheses of photosynthate efflux regulation.

The symplastic separation of maternal and embryonic tissues in

developing legume seeds requires that photosynthates must be unloaded to

the seed coat apoplast before diffusive movement to the embryo for uptake

(Thorne, 1985). The pathway of photosynthate movement from the phloem

sieve tubes to the seed coat apoplast has been examined by anatomical

evaluation of symplastic continuity in Phaseolus vulgaris, and sufficient

plasmodesmatal frequency was found to sustain the hypothesis that

photosynthates move throughout the tissues of legume seed coats before

unloading to the apoplast (the symplastic unloading route hypothesis; Offler

and Patrick, 1984). The symplastic unloading route hypothesis has also

received support from a variety of studies involving tracer kinetics (Patrick and

McDonald, 1980), mathematical modelling (Murphy, 1986), and empty seed

coat experiments (Patrick et aI., 1986; Patrick, 1987; Grusak and Minchin,

1988; Offler et aI., 1989). An alternative to the symplastic unloading route

hypothesis is that photosynthates are unloaded directly from the phloem to the

seed coat apoplast, without passage through the symplast of other cells of the

seed coat (the apoplastic unloading route hypothesis) (Wolswinkel, 1987b).

Although the greatest body of evidence supports the symplastic unloading

route hypothesis, there has been no definitive confirmation of this pathway.

During seed coat perfusion, materials may be eluted from the apoplast

or the symplast of the seed coat, and photosynthates imported via the phloem

may move into the bathing solution either directly from the phloem, or after

passage through the symplast of the seed coat. In order to clarify discussion of

93

photosynthate movement into and out of perfused seed coats, we will

distinguish between the following terms, as proposed by Grusak and Minchin

(1988); phloem import: the movement of photosynthates into the seed coat

via the phloem; phloem unloading: the movement of assimilates from the

phloem to the apoplast or the symplast of the seed coat; and seed coat

unloading: the movement of photosynthates from the seed coat symplast to

the apoplast. Steady state labelling with 14C02 was used in this study because

the relatively stable tracer fluxes produced by this method permit

straightforward analysis of photosynthate import to developing legume seeds

(Geiger and Shieh, 1988, Chapter 2). The potential disruption of the

mechanisms of phloem import and unloading by the surgical modification and

perfusion of seeds has recently been reassessed (Minchin and Thorpe, 1989;

Chapter 2), and will be considered here.

MATERIALS AND METHODS

Plant preparation, steady state labelling, and seed coat perfusion.

A determinate variety of Phaseolus vulgaris (cv. Redkloud) was grown in the

greenhouse, and individual plants were prepared for steady state labelling and

perfusion experiments 15 to 25 days after flowering. The procedure used for

steady state labelling was described in Chapter 2. In summary, steady state

labelling was initiated between 8:00 and 9:30 A.M. by sealing the top leaf in the

leaf chamber within 9 min of turning on a 1000 W metal halide lamp, which

provided approximately 1000 J,Lmol photons m-2 s-1 to the labeled leaf.

Approximately 20 min after beginning steady state labelling, "empty" seed

coats were prepared by the following protocol: 1) an incision was made

through the dorsal suture of the pod, removing roughly 45% (by fresh weight)

94

of the distal sides of two adjacent seeds, and leaving at least one intact seed

on either side of the incision, 2) the surface of the incision was smeared

liberally with silicone stopcock grease, 3) the embryos of the cut seeds were

lifted out of the seed coat in one piece using a weighing spatula, and 4) the

attached seed coat "cups" were rapidly flushed with distilled water to remove

any adhering cotyledon fragments. To determine the effect of funicular

attachment on sugar efflux, one of the two empty seed coats was selected at

random and detached from the pod by gently prying up the seed coat and

breaking the funiculus. Funicular breakage was observable when it occurred,

because detached seed coats moved slightly out of the incision in the pod wall,

and were easily pulled out of the pod.

Within 5 minutes of empty seed coat preparation, perfusion was

initiated. Seed coats were perfused with a buffer conSisting of 0.5 mM CaCI2,

1.0 mM KCI, 5.0 mM Mes, pH 6.0 (NaOH), and containing various

concentrations of mannitol as an osmoticum. Continuous perfusion was

accomplished through the use of a multi-channel peristaltic pump

(Cole-Parmer Co., Chicago, Il, U.S.A.) which transported perfusion solution

into the seed coat cup at a lower rate than it was removed. A constant level of

solution was maintained in the seed coat cup by withdrawal of solution

whenever it reached a point approximately 1.5 mm below the cut edge of the

seed coat. The rate of perfusion was between 75 and 175 III min-1 with a

typical rate of 100 III min-1, and varied no more than 25% during an individual

experiment. Perfusate was collected as 20 min samples using a fraction

collector. The first 20 min sample of perfusate was discarded, due to difficulty

in establishing uniform flow rates during the initial perfusion set-up period. At

the end of perfusion, seed coats were checked for attachment at the funiculus

95

by pulling on them with forceps. Data from seed coats that were accidently

detached were similar to data from intentionally detached seed coats.

However, data from seed coats that were accidently detached was not

reported.

Sugar and 14C analysis. 14C and sugar analysis of intact and

perfused seed tissues was described in detail in Chapter 2, and is summarized

below. Embryos (axis + cotyledons) of intact seeds were removed at the end

of experiments, combusted in a sample oxidizer, and the 14C02 that was

released was trapped and liquid scintillation counted. Perfused and intact seed

coats were extracted in hot (50 0 C) 80% (v Iv) ethanol and the extracts passed

through tandem cation (AG 50W-X8 resin, H + form; 200-400 mesh) and anion

(AG 1-X8 reSin, formate form; 200-400 mesh) exchange columns to produce a

neutral fraction (primarily sugars). An aliquot of the neutral column eluate was

liquid scintillation counted and the remainder was analyzed for sucrose +

glucose by sucrose inversion foltowed by the glucose oxidase/peroxidase

procedure (Berg meyer and Bernt, 1974). 14C remaining in extracted seed coat

tissues was determined by liquid scintillation counting, and was quantified as

the "non-extractable" fraction (containing primarily starch + structural

carbohydrates). The extractable 14C that was retained by the ion exchange

columns was calculated from the 14C content of the ethanol extracts and the

amount of 14C in the neutral fraction (corrected for percent recovery), and was

described as the "non-neutral" fraction (containing primarily amino acids and

organic acids).

The 14C content of aliquots from 20 minute samples of perfusate were

determined by liquid scintillation counting as described in Chapter 2 for ethanol

extracts that did not require de-colorization. The perfusate remaining after

removal of an aliquot for liquid scintillation counting was directly assayed for

96

sucrose + glucose after inversion using the glucose oxidase/peroxidase

procedure (Bergmeyer and Bernt, 1974). The difference between sugar efflux

from attached vs. detached seed coats at each time interval was calculated for

each experiment by subtracting the efflux of the detached seed coat from the

efflux of the attached seed coat within each pod.

The specific activity of sucrose + glucose in samples of perfusate or

column eluate was calculated based on 12 moles of carbon atoms per mole of

sucrose + glucose assayed, by assuming that 100% of the 14C in the sample

was sucrose + glucose, and that glucose and fructose were present at

equimolar concentrations. If the sample contained a significant quantity of 14C

in a substrate besides sucrose or glucose, the calculated sucrose + glucose

specific activity would be over-estimated, while a decrease in fructose relative

to glucose in the sample would lead to underestimation of the sucrose +

glucose specific activity.

Normalization of tracer data. The specific activity of 14C02 used for

labelling was used to normalize the tracer data of experiments which used

different 14C02 specific activities, so that comparisons between experiments

could be made. This was accomplished by multiplying all 14C and specific

activity measurements within an experiment by a dimensionless quantity

derived by inverting the specific activity of 14C02 used in the experiment, and

multiplying by the units of specific activity (GBq (mol Cr1). Thus, the highest

specific activity that could be reached should be 1.0 GBq mol-1 C; higher

specific activities were the result of the presence of 14C in compounds other

than sucrose, glucose, and fructose.

Size adjustment for perfused seeds. To facilitate comparisons

between intact and perfused seeds, a seed size correction factor based on the

sugar content of the basal portion of the seed coat was used to adjust

97

perfused seed data for the fraction of the seed that was lost due to excision.

Approximately 70% of the total amount of sugar in the seed coat was found in

the basal 55% of the seed (by fresh weight), and thus the use of a correction

factor based on the fraction of seed coat sugar lost by excision produced

perfused seed data that was 27% lower in magnitude than data adjusted using

a correction factor based on seed fresh weight. Given that neither the seed

fresh weight nor the seed coat sugar content correction factors were

completely accurate predictors of perfused seed coat transport capacity, the

seed coat sugar content' correction factor was used for these experiments in

order to minimize the possibility of overestimating perfused seed data in

comparisons with intact seeds. It is unlikely that perfused seed data was

underestimated by more than 30% because the correction factor based on

seed fresh weight probably overestimated the fraction of the total seed coat

tissue remaining after excision and yielded perfused seed data only 27%

greater in magnitude than the data adjusted using the seed coat sugar content

correction factor.

Sucrose uptake experiments. In the morning (between 10:00 AM and

1 :00 PM) seeds were removed from pods, and the seed coats excised by

making a longitudinal incision all the way around the seed coat, through the

hilum and funicular region. Excised seed coat pairs (one pair from each seed)

were enclosed in cheesecloth bags and placed in beakers containing solutions

of the same composition as those used for seed coat perfusions. The seed

coats remained in sucrose-free perfusion solution for a period of one hour

before they were removed from the beakers, blotted, and plunged into a

. solution of identical composition to that used for the initial elution, but

containing 200 I'M 14C-sucrose with a specific activity of 4.59 GBq mol-1 C

98

(Sigma, St. Louis, MO, U.S.A.). Seed coats were removed from the 14C_

sucrose uptake solution after defined time periods and were quickly blotted,

dipped in an unlabeled rinse solution, blotted again and placed in a second

unlabeled rinse solution for a five minute rinse period. Following the second

rinse, the seed coats were blotted and snap frozen in liquid nitrogen. The

concentration of mannitol in the solutions was varied to produce the different

treatments, but was the same for the elution, uptake, and both rinse periods of

each treatment.

For analysis of 14C, seed coats were quickly thawed, removed from the

cheesecloth bags, and dried in an oven at 75°C. Following drying, seed coats

were oxidized and counted as described previously for embryos in Chapter 2.

Uptake was approximately linear for the first 25 min of uptake, and the 5 min

rinse did not completely remove apoplastic sucrose, because 14C uptake

could not be extrapolated to zero at zero time (data not shown). Sucrose

uptake rates were calculated by subtracting the total seed coat 14C content

after a 5 min uptake period from the 14C content after an 18 min uptake period.

Uptake rate calculation from the difference in tracer accumulation at two

different uptake periods eliminates the need for a complete apoplastic washout

for the prediction of uptake into the symplast of a tissue.

Statistical methods. Analysis of covariance (Snedecor and Cochran,

1980) was used to quantify and adjust for plant to plant variability in seed coat

sugar (sucrose + glucose) content, 14C partitioning, and sugar specific activity

data collected during seed coat perfUSion experiments. Analysis of covariance

makes use of internal controls (or "covariates") that are uninfluenced by

treatments to provide a measure of the inherent variability that exists in a given

parameter from experiment to experiment. In the present study, the values of

99

14C, sugar, and sugar specific activity measured for the intact seed coats in

the pod (the internal controls, or covariates), were used to determine whether

there was a linear relationship between the data from intact seeds (the

covariates) and the values measured for the perfused seeds within the same

pod (the treatments). Covariance used intact seed coat data for the same 14C

fraction as presented for perfused seed coats, and covariance for 14C in

perfusates was calculated using the seed coat neutral fraction. When there was

a statistically significant linear relationship between intact seed measurements

(covariates) and perfused seed measurements (treatments), then the plant to

plant variability quantified by the measurements from intact seeds could be

used to adjust the values measured for perfused seeds. The final products of

covariance analysis were the "adjusted treatment means", which were

equivalent to the values that would be produced by the treatments had there

been no variability from plant to plant. Fisher's protected LSD (Snedecor and

Cochran, 1980) was used to determine whether there were statistically

significant differences between treatment means, with or without the

application of covariance adjustment.

RESULTS

Sugar efflux from perfused seed coats. There were clear differences

between the kinetics of sugar efflux from attached and detached seed coats;

sugar efflux from detached seed coats declined rapidly over time (Fig. 3.1 B),

while the efflux of sugar from attached seed coats decreased more slowly (Fig.

3.1A). The time course of detached seed coat sugar elution approximated a

tissue washout, in that there was a continuous exponential decrease in efflux

over time. The relatively slow decline in sugar efflux from attached seed coats

100

Figure 3.1 Sugar (sucrose + glucose) efflux from perfused seed coats. A,

efflux from attached seed coats; B, efflux from detached seed coats; C,

difference between sugar efflux from attached vs. detached seed coats. Data

for each mannitol concentration are the mean of three experiments, and the

LSD (p = 0.05) between the 12 hour mean efflux rates is shown. No standard

error data are presented. However, standard error tended to increase toward

the end of the perfusion period. Zero time refers to the beginning of seed coat

perfusion.

A. Attached 2~ ~ ~4'

00 .~ ood Co

101

r-.. 1

I ..c

-+-' o o U

"'0 Q) Q) rn

o

. I LSD (0.05)

B. Detached 2 0,&

E 1 ~

"'-"

~ ~

w

Mannitol Concentration o 10 mM o 300 mM ~ 500 mM

I LSD (0.05)

u C)

c. Attached - Detached + 1.0 u ::J

(f)

0.5

o ,0 \ 0 o - 0 . 5 L..:. _____ -'-___ ---'-___ ---L.-_~_----l~---...I

o 2 4 6 ·8 10 12 Time (hours)

102

may have been an outcome of the continued import of sugar into attached

seed coats via the funiculus. There was no significant difference between sugar

efflux from attached and detached seed coats for at least one hour after the

initiation of perfusion, but, two hours after perfusion began, the efflux of sugar

from attached seed coats began to increase markedly over sugar efflux from

detached seed coats (Fig. 3.1 C). After five to six hours of perfusion, sugar

efflux from attached seed coats began to decrease relative to the efflux from

detached seed coats, and after eight hours of perfusion attached seed coat

sugar efflux was not significantly greater than detached seed coat sugar efflux

(Fig. 3.1C). In summary, sugar efflux from attached seed coats was

significantly greater than from detached seed coats between two and eight

hours after the initiation of perfusion, but no significant difference between

attached and detached sugar efflux was observed during the beginning and

end of the perfusion period.

The effects of osmoticum concentration on sugar efflux from both

attached and detached seed coats were minimal, and were not statistically

significant when compared over the full 12 hours of perfusion (Fig. 3.2, Table

3.1). The kinetics of sugar efflux from attached seed coats perfused with

10 mM mannitol appeared to differ from the kinetics observed for higher

mannitol concentrations (Figs 3.1A and 3.1C), but there were no statistically

significant differences due to osmoticum concentration in either the quantity of

sugar eluted at the beginning and end of experiments, or in the slopes of the

sugar efflux curves, probably due to the high degree of variability observed in

these measurements (statistical data not shown).

The amount of sugar remaining in attached seed coats following 12

hours of perfusion was dramatically enhanced by increasing the osmoticum

concentration in the perfusion solution (Fig. 3.2, Table 3.1). At the highest

103

Figure 3.2 The effects of osmotic environment on the elution of unlabeled

sugar and 14C from perfused seed coats. All 14C and unlabeled sugar data

are presented as a percentage of the amount present in an intact seed coat.

This figure provides a graphical display of the data presented in Tables 3.1 and

3.2 for the purpose of clarifying the effects of different osmoticum

concentrations on the import and efflux of 14C and unlabeled sugar in perfused

seed coats. Tracer data were normalized to a 14C02 specific activity of 1.0

GBq mol-1 (see Materials and Methods). Bars represent the mean of three

experiments.

200%

+'

'" 0 U -c OJ

~ +' 0

'" 100% +' c .... 0

+' c OJ 0 ~

OJ Q.

0%'

- _. - - -- -- - - - - - - - - -- ~ ...

~

-~ ~~ ~~

.... ···I·II·~····· ·1·11· W···· ··1·1 ~ ......

10 rnM 300 mM 500 mM Mannitol Concentration

_ t4C

Perfusate W%JJI t4C

Seed Coat

~ ATTACHED S+G Perfusate

~ ATTACHED S+G .-. Seed Coat 0

~

~ DETACHED S+G Perfusate

V/?2 DETACHED S+G Seed Coat

TABLE 3.1 Effects of osmotic environment on sugar (sucrose + glucose) elution from seed coats dur­ing 12 hours of perfusion. Data is presented for the amount of sugar removed by perfusion (Eluted), the amount remaining in the seed coats after perfusion (Seed coat), and the total amount eluted com­bined with the amount remaining in the seed coat (Total). Each data point is the mean of three reps, and means with the same superscript are not significantly different from one another, as determined by Fisher's Protected LSD with 95% confidence. Values in parentheses are the same data expressed as a percentage of the total sugar content within an intact non-perfused seed coat. Covariance analy­sis was used only as noted below.

Concentration Attached seed coats Detached seed coats of

* Total** Mannitol Elated Seed coat Eluted Seed coat Total (mM) ("mol) ("mol) ("mol) ("mol) ("mol) ("mol)

10 11.2a (76) 4.18c (28) 15.8b (107) 6.00a (41) 2.82a (19) 8.82a (60)

300 11.1a (76) 10.4b (70) 21.3ab (145) 4.97a (34) 3.48a (24) 8.45a (58)

500 11.2a (76) 13.9a (94) 24.9a (169) 5.37a (36) 4.07a (28) 9.44a (64)

LSDO.05 4.08 3.27 6.58 2.25 2.79 3.18

* Covariance analysis was used to adjust the treatment means in this column, because the statistical significance of covariance was 91% (p = 0.09). The total sugar content of intact seed coats was used for covariance.

** The overall confidence level for the significance of differences between treatment means in this column was 89% (p = 0.117, as determined by one way analysis of covariance). For this reason the superscripted differences and the LSD presented for this column are not protected at 95% confidence, but are protected at the 89% confidence level. Covariance analysis was used to adjust the treatment means in this column even though the statistical significance of covariance was small (p = 0.410), so that the values in this column could be compared with other values for attached perfused seed coats. The total sugar content of intact seed coats was used for covariance.

.... 0 (II

106

osmoticum concentration used (500 mM), the amount of sugar remaining in

attached perfused seed coats was similar to the amount present in intact seed

coats, a result which suggests that sugar was retained in the symplast of the

seed coat as a consequence of this treatment (Fig. 3.2, Table 3.1). The sugar

contents of detached seed coats after perfusion with higher mannitol

concentrations were also higher, but there were no statistically significant

differences between osmoticum treatments (Fig. 3.2, Table 3.1). The amount of

sugar remaining in detached seed coats after perfusion was small in

comparison with attached seed coats, except when 10 mM mannitol was

present in the perfusion solution (Fig. 3.2, Table 3.1).

Net import of sugar to attached seed coats was demonstrated by the

observation that the amount of sugar eluted plus the amount retained in the

seed coat after perfusion was greater than the amount present in intact seed

coats (Fig. 3.2, Table 3.1). Higher osmoticum concentrations in the seed coat

apoplast significantly enhanced the combined total of sugar eluted from and

remaining in attached seed coats, but not in detached seed coats, which

suggests that net sugar import was enhanced by high osmoticum

concentrations (Fig. 3.2, Table 3.1).

The observation that the total amount of sugar eluted from and

remaining in detached seed coats after perfusion was 40% lower than the

sugar content of intact seed coats may be partially explained by the lack of

compensation for seed coat respiration during perfusion, which could account

for at least 7% of the intact seed coat sucrose content when integrated over the

full 12 hours of perfusion (estimated from Gifford and Thorne, 1986). In

addition, the initial 20 minutes of seed coat perfusion were not included in

sugar elution measurements, and thus the quantity of sugar eluted by

107

perfusion was underestimated by approximately 10% of the total intact seed

coat sugar content (data not shown). Estimated seed coat respiration and the

initial washout of sugar were not sufficient to account completely for the

difference between the detached seed coat sugar content (eluted + retained)

and the intact seed coat sugar content. Potential explanations for this inequality

are that respiration was greater than was estimated from data for attached

soybean seed coats (Gifford and Thorne, 1986), and that there may have been

some metabolism of sugars into unmeasured compounds during the perfusion

period.

Tracer efflux from perfused seed coats. One hour after perfusion

was initiated, the efflux of 14C from attached seed coats increased rapidly to a

maximal rate, which was followed by slowly declining rates within three to

seven hours after beginning perfusion (Fig. 3.3). The rate of tracer efflux was

significantly increased by high osmoticum concentrations during the initial four

hours of perfusion, but after this point there were no statistically significant

differences between the rates of tracer efflux at different osmoticum

concentrations, due in part perhaps to the high variability in data near the end

of the perfusions (Fig 3.3). When the quantities of 14C eluted over the entire 12

hours of perfusion were compared, no significant effect of osmoticum

concentration on tracer efflux was observed (Fig. 3.2, Table 3.2).

High osmoticum concentrations in the perfusion solution significantly

increased the 14C content of the seed coat neutral and non-neutral fractions,

and also increased the overall 14C content of the seed coat (Table 3.2). The

enhancement of seed coat 14C content by high osmoticum concentrations

appeared to parallel the increase in seed coat sugar content observed in

attached seed coats after perfusion with high mannitol concentrations (Fig.

108

Figure 3.3 14C efflux from attached, perfused seed coats. Data for each

mannitol concentration are the covariance adjusted means of three

experiments (see Materials and Methods; significance of covariance: p =

0.24). The LSD (p = 0.05) between the 12 hour mean efflux rates is shown.

Standard error data are shown at hourly intervals for the finalS hours of

perfusion to illustrate the variability present during the final hours of the

perfusion time course. Tracer data were normalized to a 14C02 specific activity

of 1.0 GBq mol-1 (see Materials and Methods). Zero time refers to the

beginning of seed coat perfusion.

c 0 .-....... 0 La ....... c Q) 0 c 0 u 0 ....... c c 0 ~

~~~ E E E 000 ..... 00

nU1

OD<l

r"'. U1 o . o ......... o Cf) .....J

109

8-' o

co

~--~--~----~--~--------~ 0 ID V N 0

( L - 4 L _lDOO pees bS>1) xnlJJ3 J-v L

TABLE 3.2 Effects of osmotic environment on 14C elution from seed coats during 12 hours of perfusion. Data is present­

ed for the 14C remaining in the seed coat after perfusion, the 14C eluted during perfusion (Eluted), the combined total of

14C eluted and 14C remaining in the seed coat (also given as a percentage of the total 14C in an intact seed), and the

average rate of 14C efflux from the seed coat during the initial 4 hours of perfusion. Each data point is the mean of three

experiments, and means with the same superscript are not significantly different from one another, as determined by

Fisher's Protected LSD with 95% confidence. Values in parentheses are the same data expressed as a percentage of the

14C content in the equivalent fraction of an intact seed coat. Covariance analysis was used to adjust the treatment means

in this table, and the statistical significance of covariance analysis is presented at the bottom of each column (see

Materials and Methods for details). Tracer data were normalized to a 14C02 specific activity of 1.0 GBq mol-1 (see

Materials and Methods).

14C remaining in seed coat Combined Combined

Concentration total total as Initial of seed coat percent of 4 h efflux

Mannitol Neutral Non-Neutr. Non-Extract. Total Eluted + eluted intact seed rate (mM) (kBq) (kBq) (kBq) (kBq) (kBq) (kBq) (%) (kBq h-1)

10 24.8b (30) 13.3b (48) 4.49a (83) 43.2c (38) 32.7a (40)* 77.6a (68)* 25.8 1.60b

300 41.2a (50) 17.1a (62) 3.13a (58) 60.3b (54) 29.0a (36) 88.0a (76) 29.3 2.58a

500 51.1a (62) 17.1a (62) 4.42a (81) 73.2a (64) 29.1a (36) 102.0a (89) 33.9 2.63a

LSDO.05 12.3 3.14 1.60 12.2 13.8 23.1 0.762

Covariance p-value 0.04 0.003 0.05 0.016 0.23 0.044 0.24

*The percent of intact seed coat data for this column was calculated using the intact seed coat neutral fraction.

.... .... 0

111

3.2). The amount of 14C in the seed coat non-extractable fraction was not

significantly modified by osmoticum concentration, so that there was no

evidence for changes in starch storage or mobilization caused by osmoticum

treatment (Table 3.2).

The total amount of 14C imported to perfused seed coats could be

quantified by combining the amount of 14C released into the perfusion solution

with the 14C content of the perfused seed coat (Fig. 3.2, Table 3.2). There was

no statistically significant effect of osmoticum on the total amount of 14C

imported to perfused seed coats. However, this was probably due to the large

variance and lack of osmoticum effects observed for the eluted portion of the

data for total seed coat 14C import (Table 3.2). The total amount of 14C

imported to perfused seed coats was only slightly lower than the amount of

14C accumulated in intact seed coats, but was approximately 70% lower than

the total amount imported to intact seeds (seed coat + embryo) (Fig. 3.2,

Table 3.2). The use of a conservative seed size correction factor may have

underestimated perfused seed 14C data by as much as 30% (see Materials

and Methods), but the large difference between intact and perfused seed 14C

import could not be fully explained by seed size underestimation. Thus, the

lowered 14C import observed in perfused seeds relative to intact seeds was

most likely due to the inhibition of photosynthate import by the empty seed

coat technique.

Sugar specific activity during seed coat perfusion. The specific

activity of sugar eluted from attached perfused seed coats continuously

increased after one hour of perfusion, although there was some evidence of a

lag in specific activity increase between 7 and 9 hours after the initiation of

perfusion (Fig. 3.4). The significance of the lag in sugar specific activity

112

Figure 3.4 Specific activity of sugar (sucrose + glucose) in perfusate from

attached seed coats. Data for each mannitol concentration are the covariance

adjusted means of three experiments (see Materials and Methods; significance

of covariance: p = 0.37). The LSD (p = 0.05) between the 12 hour mean sugar

specific activities are shown. Standard error data are also presented. Specific

activity data were normalized to a 14C02 specific activity of 1.0 GBq mol-1 (see

Materials and Methods). Zero time refers to the beginning of seed coat

perfusion.

c a .-...., o L.. ...., C OJ

o U') ...J

o c~~~

8 E E E -000 0-00 ~ f"")l.() c §OO<J ~

113

00 ~ ~ N 0 . . . . . o 0 000

N

o

N

o

(~_IOW bStJ) AlIl\llOV 0IJ!oedS 0l~+ons

114

increase between 7 and 9 hours after the initiation of perfusion is unknown. The

average specific activity of sugar eluted over the full 12 hours of perfUSion may

have been increased by low concentrations of osmoticum in the perfusion

solution, although differences between osmoticum treatments were not

significant at the 95% confidence level (Table 3.3). In addition, the slope of the

sugar specific activity time course between two and ten hours after the initiation

of perfusion was significantly increased by low osmoticum concentrations

(Table 3.3). During the final three hours of perfusion, the osmoticum

concentration of the perfusion solution did not produce statistically significant

differences in the specific activity of eluted sugar (Table 3.3), and this was

probably due to the high variability in specific activity measurements during the

final three hours of perfusion (Fig. 3.4).

The trend toward increased sugar specific activity at lower osmoticum

concentrations that was observed for eluted sugar was also visible in the data

for the sugar remaining in the seed coat, despite the lack of statistical

significance for this observation (Table 3.3). The pattern of higher sugar

specific activities at lower osmoticum concentrations was the opposite of the

decrease in seed coat sugar and 14C contents at lower osmoticum

concentrations. The specific activity of sugar remaining in perfused seed coats

was lower than in intact seed coats, regardless of osmoticum concentration

(Table 3.3).

The specific activity of sugar in seed coat perfusates was not directly

comparable with the specific activity of sugar remaining in perfused seed

coats, because perfusates contained significant quantities of 14C in

substances besides sugars that caused the specific activity measurements for

perfusates to be overestimated (see Materials and Methods).

115

TABLE 3.3 Effects of osmotic environment on the specific activity of sugar (sucrose + glucose) eluted from and remaining in perfused seed coats. Specific activity of sugar remaining in the seed coats after 12 hours of perfusion is presented in the column titled "Seed coat". Data for the specific activity of sugar eluted during perfusion is described by the average over the full 12 hours of perfusion (0-12 h), the average over the last 3 hours of perfusion (Final 3 h), and the slope of the increase in specific activity between 2 and 10 hours after initiation of perfusion (2-10 h S.A. slope). Each data point is the mean of three experiments, and means with the same superscript are not significantly different from one another, as determined by Fisher's Protected LSD with 95% confidence. Values in parentheses are the same data expressed as a percentage of the specific activity of sugar extracted from intact non-perfused seed coats at the end of experiments. Covariance analysis was used to adjust the treatment means in this table, and the statistical significance of the covariance analysis is presented at the bottom of each column. Covariance analysis used the specific activity of sugar in intact seed coat-r 4at the end of each experiment. Specilic activity data were normalized to a C02 specific activity of 1.0 G8q mol- (see Materials and Methods).

Concentration Average perfusate S.A. Perfusate of 2-10 h

Mannitol Seed coat 0-12 h* Final 3 h* S.A. slope *

(mM) (G8q mol-1) (G8q mol-1) (G8q mol-1) (G8q mol-1 h-1)

10 0.386a (81) 0.356a 0.668a 0.0589a

300 0.348a (73) 0.327ab 0.437a 0.0334b

500 0.340a (72) 0.268b 0.443a 0.0332b

LSDO.05 0.090 0.080** 0.321 0.0134

Covariance p-value 0.013 0.374 0.408 0.200

* Covariance analysis was used to adjust the treatment means in these columns even though the statistical significance of covariance was low (p > 0.10). Covariance with low significance was used to allow relative comparisons between the different columns of this table, which otherwise would not be comparable.

** The overall confidence level for the significance of differences between treatment means in this column was 91 % (p = 0.091, as determined by one way analysis of variance). For this reason, the superscripted differences and the LSD presented for this column are not protected at 95% confidence, but are protected at the 90% confidence level.

116

Sucrose uptake by excised seed coats. The rate of sucrose uptake

by excised seed coats was determined after one hour of elution in a buffer

identical in composition to that used to perfuse attached and detached seed

coats. The sucrose concentration used for seed coat uptake experiments was

200 I'M, a concentration that was chosen for its similarity to the concentration

of sucrose + glucose in the perfusate of seed coats after one hour of perfusion

(data not shown). The rate of excised seed coat sucrose uptake after one hour

of elution (Fig. 3.5) was so small as to be negligible in comparison with the rate

of sugar efflux from detached seed coats after one hour of perfusion (Fig.

3.18). In addition, osmoticum concentration had no effect on the rate of

sucrose uptake into excised seed coats (Fig. 3.5).

DISCUSSION

The effects of perfusion on photosynthate import and efflux from

seed coats. The observation that 14C import to perfused seeds was

significantly lower than 14C import to intact seeds contrasted with previous

studies in which no significant differences were observed between perfused

and intact seed 14C import after pulse-labelling (Thorne and Rainbird, 1983;

Wolswinkel and Ammerlaan, 1983; Wolswinkel and Ammerlaan, 1984). The

tracer fluxes produced by pulse-labelling are not stable over time, and for this

reason, the accumulation of pulse-labeled tracer in sink tissues such as seeds

does not necessarily reflect the rate of photosynthate import over time (Zierler,

1981; Thorne, 1985; Geiger and Shieh, 1988). The relatively stable tracer fluxes

produced by steady state labelling allow reliable quantitative analyses of tracer

accumulation rates in seeds (Chapter 2), and thus the observed inhibition of

117

Figure 3.5 Sucrose uptake by detached seed coats. Uptake at each mannitol

concentration represents the mean of five replications, and error bars are 95%

confidence intervals.

118

N a o 0 . . a a

a a L{) ",-... ~ E c:

00 0:';:; r")O

L... ..fo-J c: Q) o c:

00 au ~-o

..fo-J

c: c: o

o~

~ _loOO pees lown) e>1oldn eso.Jons

119

tracer import into perfused seeds relative to intact seeds provides strong

evidence for the inhibition of photosynthate import by application of the empty

seed coat technique.

At least part of the inhibition of photosynthate transport in perfused

seeds may have been caused by surgical removal of the distal halves of seeds,

because this procedure reduced photosynthate transfer from the seed coat to

the embryo (Chapter 2). The inhibition of tracer movement out of perfused

seed coats relative to intact seed coats was demonstrated by the observation

that the proportion of total perfused seed 14C (seed coat + eluted) that

remained in the seed coat was much larger than the proportion of total intact

seed 14C (seed coat + embryo) that remained in the seed coat (Table 3.2,

Chapter 2). This data confirms the observations of Minchin and Thorpe (1989)

who used multiple pulse-labelling with 11 C to demonstrate that during a six

hour perfusion period, the movement of tracer from the seed coat to the

perfusion solution decreased considerably, with a simultaneous decline in total

tracer import to perfused seeds. In summary, the empty seed coat technique

inhibited both photosynthate import to the seed and photosynthate unloading

from seed coats, so that surgically modified and perfused seed coats were not

identical to seed coats in vivo.

Despite the observation that tracer import into perfused seeds was

inhibited relative to intact seeds, there were significant fluxes of 14C_

photosynthate into and out of perfused seed coats (Fig. 3.2, Table 3.2). In

addition, the net phloem import of sugar to attached perfused seed coats was

demonstrated by the observation that the total amount of sugar eluted from

and remaining in perfused seed coats was significantly greater than the

amount originally present in intact seed coats (Fig. 3.2, Table 3.1). Although

120

photosynthate import and unloading in perfused seed coats was reduced in

comparison with intact seed coats, the sugar and 14C-photosynthate fluxes

into and out of perfused seed coats were quantitatively significant, and may

reflect the same mechanisms that regulate photosynthate unloading in vivo.

Properties of the phloem unloading pathway in perfused seed

coats. The apoplastic movement of photosynthates between the phloem and

other cells in perfused seed coats was probably greatly reduced by the

continuous washout of apoplastic solutes into the perfusion solution. This

indicates that sugar accumulation within perfused seed coats was primarily

confined to the phloem and tissues that were symplastically connected with the

phloem. The minute volume of the phloem in comparison to the volume of

other tissues in the seed coat suggests that a significant proportion of the total

amount of sugar in intact seed coats was present in tissues other than the

phloem (Gifford and Thorne, 1986). The high sugar content of attached

perfused seed coats in comparison with detached seed coats (Fig. 3.2, Table

3.1) may thus provide evidence for the hypothesis that the sugar imported by

the phloem in attached seed coats was transported symplastically into other

tissues in the seed coat. This is because the high sugar content of attached

seed coats relative to detached seed coats was due to the net import of sugar

via the phloem, and the approximately equal sugar contents of attached and

intact seed coats when SOD mM mannitol was present in the perfusion solution

implies that sugar had moved into cells outside of the phloem. An alternative

hypothesis is that a the higher sugar content of attached seed coats resulted

from a reduction in sugar efflux from the bulk symplast of attached seed coats

relative to efflux from the symplast of detached seed coats. These two

hypotheses can not be distinguished on the basis of unlabeled sugar

measurements alone.

121

Additional evidence for the symplastic movement of sugars from the

phloem throughout the seed coat was provided by the observation that the

specific activity of sugar in perfused seed coats was only 30% lower than in

intact seed coats (Table 3.3). Because the apoplastic movement of

photosynthate between the phloem and other cells in the seed coat was

probably negligible during perfusion, only the symplastic movement of sugar

throughout the seed coat could have allowed the majority of seed coat sugar to

exchange with the labeled sugar imported by the phloem. If a significant

proportion of the sugar in the seed coat did not exchange with the,labeled

sugar imported by the phloem, a dramatic reduction in perfused seed coat

sugar specific activity would be expected. The observed reduction in perfused

seed coat sugar specific activity was relatively small in comparison to the

specific activity reduction in cut seed coats relative to intact seed coats

(Chapter 2), particularly when considering that 14C import was inhibited to a

greater extent in perfused seeds than in cut seeds. Thus, under conditions in

which the sugar contents of intact and perfused seed coats were similar

(500 mM mannitol treatment), the relatively small difference between the sugar

specific activities of intact and perfused seed coats required the symplastic

movement of sugars from the phloem throughout the seed coat.

The observation of symplastic phloem unloading in perfused seed coats

does not rule out the possibility of direct unloading from the phloem to the seed

coat apoplast. Some degree of direct phloem unloading to the apoplast

probably occurs in parallel with symplastic unloading (Wolswinkel et aI., 1986,

Wolswinkel, 1987b), but the symplastic movement of photosynthates

throughout the seed coat implies that the majority of photosynthate unloading

to the apoplast takes place across the combined membrane surface areas of

many cells within the seed coat as proposed by Offler and Patrick (1984).

122

General characteristics of photosynthate efflux from perfused seed

coats. The hypothesis that sugar efflux from the seed coat was at a steady

state with the rate of sugar import via the phloem was utilized by Gifford and

Thorne (1986) as the basis for phloem unloading rate determination. Sugar

eluted during the initial 2-3 hours of perfusion was attributed to the washout of

sugar from the seed coat apoplast and symplast, while sugar efflux after this

period was said to reflect the rate of continued phloem import and unloading to

the seed coat apoplast (Gifford and Thorne, 1986; Ellis and Spanswick, 1987).

The component of attached seed coat sugar efflux that was attributed to

steady state phloem unloading by Gifford and Thorne (1986) accounted for

about half of the in vivo rate of sugar import into intact seeds of Glycine max.

Due to the lack of phloem import into detached seed coats, the

difference between sugar efflux from attached versus detached seed coats

should reflect the influence of phloem import on sugar efflux from the seed

coat. The maximum difference between attached and detached seed coat

sugar efflux was 1.0 J£mol sucrose + glucose seed-1 h-1 (Fig. 3.1 C), while the

estimated in vivo rate of sucrose accumulation within intact seeds was

probably greater than 2.1 J£mol sucrose seed-1 h-1 (Chapter 2). Thus, the

difference between attached and detached seed coat sugar efflux accounted

for less than half of the unloading rate that would be required to support seed

growth in vivo. The low sugar import rates predicted from perfused seed coat

sugar efflux measurements may be explained by the observed inhibition of

photosynthate import to perfused seeds relative to intact seeds.

The difference in the rate of sugar elution of attached versus detached

seed coats could be used to calculate the proportion of the seed coat sugar

that was involved in the flux between phloem import and unloading to the

perfusion solution (Le. the turnover rate). This calculation demonstrated that

123

approximately 7% of seed coat sugar passed through the seed coat from the

phloem to the perfusion solution in one hour when 500 mM mannitol was used

for perfusion, while perfusion with a 10 mM mannitol solution increased the

proportion of seed coat sugar exchanged between phloem import and

unloading to approximately 24% of the total seed coat sugar content per hour.

The use of sugar efflux analysis for the measurement of photosynthate

unloading from perfused seed coats was complicated by the lack of clearly

distinguishable steady state efflux components in these experiments. If the

phloem import of sugar was constant during the perfusion period, and if the

only difference between attached and detached sugar efflux was due to

continued phloem import and unloading in attached seed coats, then the

difference between sugar efflux from attached and detached seed coats should

be fairly constant over time. Data presented in Figure 3.1 C demonstrated that

this was not the case, and the lack of a significant difference between attached

and detached seed coat sugar efflux at the beginning and end of the perfusion

period may indicate that phloem import and/or unloading was inhibited at the

beginning and end of the perfusion period. Additional evidence for the

inhibition of photosynthate unloading during the final hours of perfusion was

provided by the observation that the efflux of 14C-photosynthate decreased

near the end of the perfusion period (Fig. 3.3). Alternatively, the lack of a

constant difference between the sugar effluxes of attached and detached seed

coats may indicate that sugar washout from the seed coat symplast was not

kinetically distinct from the sugar fluxes caused by phloem import and

unloading to the apoplast.

Regardless of whether or not a kinetically distinct phloem unloading

component of sugar efflux exists, the kinetics of sugar efflux from perfused

seed coats were complex, and there was no reliable method for the

124

quantitative determination of phloem unloading rates based on the analysis of

unlabeled sugar efflux measurements. As observed by Wolswinkel (1987a),

and Gifford and Thorne (1986), sugar efflux measurements were not useful for

predicting phloem import and unloading until at least two hours after perfusion

was initiated, and the results of the current study demonstrate that after

approximately 8 hours of perfUSion, sugar efflux was not significantly enhanced

by seed coat connection with the phloem. Therefore, sugar efflux

measurements obtained from perfused seed coats either before two hours or

after eight hours of perfusion were not useful for the determination of phloem

import and unloading. Although the quantitative significance of the difference in

sugar efflux from attached versus detached seed coats remains unclear, sugar

efflux between two and eight hours after the initiation of seed coat perfusion

was probably influenced by phloem import and unloading during this period.

For this reason, sugar efflux between two and eight hours after beginning

perfusion may provide a useful measure for comparisons between the effects

of different treatments on phloem import and unloading in perfused seed coats.

Turgor effects on photosynthate import and unloading in perfused

seed coats. The net import of sugar and 14C to attached perfused seed coats

was enhanced by high osmoticum concentrations in the seed coat apoplast

(Fig. 3.2, Tables 3.1 and 3.2). This indicates that phloem import was enhanced

by decreased cell turgor in the seed coat, because the primary effect of

osmoticum treatment is the modification of cell turgor in perfused seed coats

(Patrick, 1984). Thus, the results of this study confirm that phloem import to

perfused seed coats is enhanced by low cell turgor, as reported by previous

researchers (Wolswinkel and Ammerlaan, 1984, 1986; Wolswinkel et aI., 1986;

Minchin and McNaughton, 1986; Ellis and Spanswick, 1987; Grusak and

Minchin, 1988).

125

Several studies have reported the stimulatory effects of high osmoticum

concentrations on photosynthate efflux from seed coats, and have discussed

these observations with respect to phloem import stimulation by low cell turgor

in the seed coat (Wolswinkel and Ammerlaan, 1984, 1986; Wolswinkel et aI.,

1986; Minchin and McNaughton, 1986; Ellis and Spanswick, 1987; Grusak and

Minchin, 1988). High osmoticum concentrations in the seed coat apoplast

increased the initial efflux of 14C-photosynthate from perfused seed coats

(Table 3.2), and this data agrees with the conclusions of the previous studies.

In contrast with these studies, however, no long-term effects of cell turgor on

net sugar or 14C-photosynthate efflux were observed in these experiments

(Fig. 3.2, Tables 3.1 and 3.2), or in the investigations of Thorne and Rainbird

(1983) and of Gifford and Thorne (1986). The results of the current study

indicate that, although the net phloem import of sugar and 14C-photosynthate

was enhanced by low cell turgor, increased phloem import did not directly lead

to increases in the efflux of imported materials from the seed coat.

The lack of direct coupling between net photosynthate import and net

photosynthate efflux that was observed in these and other experiments may be

explained by the hypothesis that net photosynthate import via the phloem was

increased by low cell turgor, whereas photosynthate efflux from the seed coat

was decreased by low cell turgor (Wolswinkel et aI., 1986; Grusak and Minchin,

1988; Minchin and Grusak, 1988). Patrick (1984) attributed the turgor­

sensitivity of photosynthate efflux to the action of a turgor-sensitive sugar

carrier acting as a turgor homeostat, which would increase photosynthate

efflux from the seed coat when cell turgor was increased above a "turgor set

point" (Patrick, 1984; Patrick et aI., 1986). Photosynthate efflux regulation by a

turgor homeostat mechanism could provide a balance between photosynthate

126

import and unloading within intact seed coats, but the enhancement of net

photosynthate efflux by perfusion may have induced a relatively constant efflux

of photosynthates at the expense of photosynthate retention in the seed coat

symplast. The significant depletion of symplastic sugars and other

photosynthates from detached seed coats by long-term (12 hour) perfusion

may be responsible for the lack of turgor-sensitivity of efflux in detached seed

coats. However, the maintenance of relatively high sugar contents in attached

seed coats perfused with high osmoticum concentrations would provide a

ready source of photosynthates for efflux to the apoplast. Under conditions of

increased phloem import to the seed coat caused by low cell turgor, the

retention of symplastic sugars was similar to that observed in intact seed coats,

while high cell turgor conditions were associated with the net loss of sugars

from the seed coat symplast (Fig. 3.2, Table 3.1). Thus, the existence of a

turgor-sensitive efflux control mechanism may explain the observation that net

photosynthate import into perfused seed coats was enhanced by low cell

turgor without changes in net photosynthate efflux.

The regulation of photosynthate efflux by the action of a turgor

homeostat mechanism would lead to relatively constant concentrations of

photosynthates in the apoplast of intact seed coats and embryos because, at a

steady state, the increase in net phloem import by low osmotic potentials in the

apoplast would be balanced by the decreased efflux of photosynthates

(Patrick, 1984). Reduction of apoplastic photosynthate concentrations by

increased rates of uptake by the embryo would in turn lead to increased

photosynthate efflux, thus providing coordination of seed coat unloading with

photosynthate uptake by the developing embryo.

127

The role of apoplastic retrieval in photosynthate unloading from

perfused seed coats. The hypothesis that photosynthate efflux from the seed

coat was regulated by the activity of a turgor-sensitive apoplastic retrieval

mechanism provides an alternative to the turgor homeostat mechanism of

photosynthate efflux regulation (Wolswinkel and Ammerlaan, 1986; Wolswinkel

et al. 1986). A turgor-sensitive apoplastic retrieval mechanism could increase

the net efflux of photosynthates by decreasing the net retrieval of

photosynthates from the apoplast in response to high cell turgor (Wolswinkel

and Ammerlaan, 1986; Wolswinkel et al. 1986).

There was no significant effect of osmoticum concentration on the rate

of sucrose retrieval from the apoplast of excised seed coats in these

experiments, and the rate of sucrose uptake into excised seed coats was

negligible compared with the rate of sugar efflux from perfused seed coats. It

must be noted that the short uptake period used in this study probably

minimized sucrose uptake into the vacuole, and thus the data presented in

Figure 3.5 primarily represent sucrose uptake across the plasmalemma. The

results of the current study verify the absence of a turgor-sensitive apoplastic

retrieval mechanism in perfused seed coats as was earlier reported by Patrick

et al. (1986), who also demonstrated that sucrose uptake into the vacuoles of

the seed coat symplast was not turgor-sensitive. These results do not preclude

the possibility that significant retrieval of photosynthates from the apoplast,

turgor-sensitive or otherwise, may occur in intact seed coats, because the

concentration of sugars and other photosynthates are very high in the apoplast

of intact seed coats in comparison with perfused seed coats (Patrick and

McDonald, 1980; Gifford and Thorne, 1985).

128

Tracer kinetics in perfused seed coats. Under steady state

conditions in which sugar import to the seed coat was equal to the net efflux of

sugar from the seed coat, the enhancement of phloem import by low cell turgor

would be expected to increase the specific activity of seed coat sugar. The

opposite was observed in these experiments, because high osmoticum

concentrations in the seed coat apoplast produced low specific activities in the

sugar eluted from and remaining in perfused seed coats (Table 3.3).

The sugar content of attached seed coats was reduced below the

amount originally present in intact seed coats by perfusion with low

concentrations of osmoticum, while the sugar content of seed coats perfused

with high osmoticum concentrations was not. The low symplastic sugar

content of seed coats perfused with low osmoticum concentrations probably

reduced the amount of unlabeled sugar in the seed coat, and the small

symplastic sugar pool remaining in the seed coat could rapidly equilibrate with

the relatively high specific activity sugar imported by the phloem. Thus, the

increased specific activity of sugar remaining in seed coats perfused with low

osmoticum concentrations would be expected.

The greatest loss of sugar from perfused seed coats occurred during

the initial hours of perfusion, which indicates that the sugar content of the seed

coat symplast was larger during this period, and that the efflux of unlabeled

sugar was also relatively high (Gifford and Thorne, 1986; Ellis and Spanswick,

1987). In addition, the specific activity of sugar imported by the phloem was low

during the early hours of steady state labelling (Chapter 2). Thus the specific

activity of sugar eluted during the initial hours of seed coat perfusion could not

be strongly influenced by the rate of phloem import, because the efflux of low

specific activity seed coat sugar was probably great enough to dilute the

129

labeled sugar imported by the phloem. Therefore, the lack of significant

osmoticum-induced changes in the specific activity of sugar eluted during the

initial hours of perfusion could be explained (Fig. 3.4). After the initial hours of

perfusion however, the sugar eluted from seed coats perfused with low

osmoticum concentrations was at a higher specific activity (Fig. 3.4), which

was probably the result of the higher specific activity of sugar in the symplast of

seed coats perfused with low osmoticum concentrations, as described in the

preceding paragraph.

CONCLUSIONS

The empty seed coat technique significantly inhibited both the import

and efflux of photosynthates from perfused seed coats relative to intact seeds.

However, phloem import and unloading did occur in perfused seed coats, and

the control of these mechanisms may reflect the processes that occur in intact

seeds. Taken together with the data from previous studies (Wolswinkel and

Ammerlaan, 1984, 1985a, 1986a, 1988; Wolswinkel et aI., 1986; Minchin and

McNaughton, 1986; Ellis and Spanswick, 1987; Grusak and Minchin, 1988), the

current experiments provide virtually conclusive evidence that low cell turgor

within perfused seed coats enhances net phloem import. The symplastic

movement of photosynthates throughout the seed coat symplast was

investigated, and the symplastic pathway of phloem unloading (Offler and

Patrick, 1984) was supported for experiments with perfused seed coats. There

was no evidence for a turgor-sensitive apoplastic retrieval mechanism in

perfused seed coats as proposed by Wolswinkel and Ammerlaan (1986). The

partitioning of photosynthates between retention in the seed coat and

130

unloading to the apoplast was turgor-sensitive, and could be explained by the

turgor homeostat model of photosynthate unloading as described by Patrick et

al. (1986).

CHAPTER 4

Rapid Changes in Photosynthate Unloading Induced by

Osmoticum and EDTA

INTRODUCTION

The empty seed coat technique facilitates the measurement of

photosynthates released from maternal seed tissues following the surgical

removal of the embryo and the addition of a trapping solution or agar (Thorne

and Rainbird, 1983; Wolswinkel and Amerlaan, 1983; Patrick, 1983; Porter et

aI., 1985). Sucrose is eluted from perfused "empty" seed coats in greater

quantities than any other photosynthate, and many studies have focused on

the release of this compound (Thorne and Rainbird, 1983; Wolswinkel and

Amerlaan, 1983; Patrick, 1983; Thorne, 1985; Wolswinkel, 1988). The seed

coat and embryo of developing legume seeds are symplastically separated,

and thus photosynthates must be released to the seed coat apoplast where

photosynthates move to the embryo for uptake (Thorne, 1985). The pathway of

photosynthate movement from the phloem to the seed coat apoplast appears

to involve symplastic transport of photosynthates throughout the seed coat

before efflux to the apoplast (Patrick and McDonald, 1980; Offler and Patrick,

1984; Murphy, 1986; Patrick et aI., 1986; Patrick, 1987; Grusak and Minchin,

1988; Offler et aI., 1989; Chapter 3). However, the possibility that

photosynthates may also be unloaded directly from the phloem to the seed

coat apoplast cannot be entirely ruled out, at least in intact seeds (Wolswinkel,

1987b; Chapter 3).

EDTA and EGTA have been used for the collection of phloem sap from

cut petioles and other tissues, because these chelators promote high rates of

phloem exudation over prolonged periods (Groussol et aI., 1986). Higher

131

132

concentrations of EDTA (and EGTA) appear to produce a general "leakiness" in

the cortical cells of exuding petioles, and at 15 mM EDTA (the concentration

used in the current study and in Wolswinkel, 1987a), a significant proportion of

the sugars exuded from cut petioles were the result of leakage from cells other

than the phloem (Groussol et aI., 1986). Thorne and Rainbird (1983) measured

photosynthate efflux from soybean seed coats into agar traps containing

15 mM EGTA or EDTA and found that EGTA (or EDTA) stimulated the efflux of

pulse-labeled 14C-photosynthates into agar traps, apparently at the expense of

14C retention in the seed coat. The results of EGTA treatment were said to be

caused by massive phloem leakage and a general increase in membrane

permeability. The lack of a significant EGTA-induced increase in 14C import to

modified seeds (seed coat + agar) may indicate that EGTA treatment did not

specifically stimulate phloem import to the seed.

Hanson (1986a) determined that sucrose efflux from soybean seed

coats into a trapping solution was increased by 20 mM EDTA more than 2 mM

EDTA, and also demonstrated that EDTA-induced sucrose effluxes increased

over time relative to a solution without EDT A. The quantity of sucrose released

in the presence of 20 mM EDTA did not depend on seed coat area, while

sucrose release in the presence of 2 mM EDTA was more closely related to

seed coat area. Perfusion with 200 mM sorbitol and 2.5 mM EDTA produced a

sucrose efflux that was linearly dependent on seed coat area. The dependence

of sucrose efflux on seed coat area was used as a measure of whether efflux

was related to sucrose import to the seed coat via the phloem, although the

amount of sucrose present in the seed coat would also be expected to depend

on seed coat area, regardless of phloem import. Thus, the dependence of

sucrose efflux on seed coat area was not a clear indicator that sucrose efflux

133

was primarily due to phloem unloading of sucrose, and it appears that EDTA

treatment may have induced some degree of general leakiness in this system.

Wolswinkel (1987a) demonstrated that the efflux of unlabeled solutes (including

sucrose) from attached Vicia faba seed coats treated with 400 mM mannitol +

15 mM EDTA was increased over efflux from a 400 mM mannitol control, and

showed that the efflux-enhancing effects of EDTA increased over time. EDTA

treatment produced similar effects in excised seed coats, though the amount of

efflux was lower. Wolswinkel (1987a) indicated that both phloem import and

photosynthate release were stimulated by EDTA, but the effect of EDTA on net

phloem import was not directly demonstrated.

14C-photosynthate efflux from excised seed coats of Phaseolus vulgaris

is rapidly enhanced when the perfusion solution is changed from a 100 mM

mannitol solution to a solution without mannitol (Patrick, 1984; Patrick et al.,

1986). The cause of changes in 14C efflux was proposed to be the stimulation

of a turgor-sensitive efflux mechanism in cells near the inner surface of the

seed coat by a less negative osmotic potential in the apoplast (high cell turgor)

(Patrick et aI., 1986). In attached soybean seed coats the efflux of

11 C-photosynthate was clearly stimulated by changing from a low to a high

mannitol concentration in the perfusion solution (200 to 400 to 1000 mM), and

was also decreased by changing from a high to a low concentration of

mannitol (300 to 0 mM; Minchin and McNaughton, 1986). This was explained

by an increase in phloem import at high osmoticum concentrations due to the

lowering of phloem turgor (Minchin and McNaughton, 1986). In addition, direct

monitoring of 11 C import to perfused soybean seed coats demonstrated that

changing from a low to a high osmoticum concentration in the perfusion

solution increased phloem import rapidly, while tracer release to the perfusion

134

solution increased more slowly (Grusak and Minchin, 1988). These data

support the hypothesis that the increase in tracer efflux caused by high

osmoticum treatment was secondary to, and dependent on, the increase in

phloem import (Grusak and Minchin, 1988). Thus, the presence of high

concentrations of osmoticum in perfusion solutions may lead to increased

photosynthate retention in seed coats relative to release from the seed coat to

the perfusion solution (Minchin and Grusak, 1988; Chapter 3).

Changing the mannitol concentration in the perfusion solution from a

high (SOO mM) or intermediate concentration (100 or 200 mM) to a low

concentration (10 mM), or changing from a low or intermediate concentration

to a high concentration produced rapid increases in unlabeled sugar (sucrose

+ glucose) efflux from attached soybean seed coats (Ellis and Spanswick,

1987). The observation that sugar efflux was increased to a greater extent and

for a longer period following the change from a low to a high osmoticum

concentration than after the change from a high to a low osmoticum

concentration may indicate that high osmoticum concentrations generate long

term efflux stimulations by increasing phloem import, while low osmoticum

concentrations produce a transient increase in efflux by the direct stimulation of

a turgor-sensitive efflux control mechanism (Ellis and Spanswick, 1987).

The source of photosynthates released from attached perfused seed

coats is not easily determined, because photosynthate efflux may occur

directly from the phloem or the seed coat symplast, or by symplastic

movement from the phloem throughout the seed coat symplast before efflux to

the perfusion solution (Patrick et aI., 1986; Gifford and Thorne, 1986; Grusak

and Minchin, 1988; Chapter 3). The purpose of this study is to determine

whether the efflux of sugar and 14C-photosynthates from different sources

135

within perfused seed coats may be differentially stimulated by treatment with

EDTA or by rapid changes of osmoticum concentration in the perfusion

solution. Steady state labelling with 14C02 was used to simplify analysis of

tracer import and efflux, because tracer fluxes produced by this method are

relatively stable and predictable (Geiger and Shieh, 1988; Chapter 2; Chapter

3).

MATERIALS AND METHODS

Plant preparation, steady state labelling, and seed coat perfusion.

Plants of Phaseolus vulgaris (cv. Redkloud) were grown in the greenhouse,

and were prepared for steady state labelling and perfusion experiments 15 to

25 days after flowering as described in Chapter 2 and summarized below.

Steady state labelling was initiated by enclosing the top leaf in a leaf chamber

within 9 min of leaf illumination (1000 J.'mol photons m-2 s-1). Approximately

20 min after beginning steady state labelling, "empty" seed coats were

prepared by the protocol described in Chapter 3. Within 5 minutes of empty

seed coat preparation, seed coats were perfused with a solution containing

0.5 mM CaCI2, 1.0 mM KCI, 5.0 mM Mes, pH 6.0 (NaOH). Various

concentrations of mannitol (as an osmoticum) and 15 mM EDTA were also

present in the perfusion solution where indicated. A constant level of solution

was maintained in the seed coat cup by the slow addition and rapid withdrawal

of solution by a peristaltic pump. The typical rate of perfusion was

100 J.'L min-1, and varied no more than 25% during an individual experiment.

Perfusate was collected as 10 min samples using a fraction collector and

perfusate collected during the first 20 minutes of perfusion was discarded due

to the initial difficulty in establishing uniform flow rates. To study the effects of

136

changes in mannitol concentration and the presence of EDTA, the perfusion

solution was changed at the input to the peristaltic pump. The lag time between

solution changes at the pump input and the appearance of the new perfusion

solution at the fraction collector was approximately 7 minutes. At the end of

perfusion, seed coats were checked for attachment at the funiculus by pulling

on them with forceps, and data from detached seed coats was not reported.

Sugar and 14C analysis. 14C and sugar analysis of intact and

perfused seed tissues was described in detail in Chapters 2 and 3, and is

summarized below. The 14C content of embryo (axis + cotyledons) from intact

seeds was determined at the end of experiments by liquid scintillation counting

of the 14C released following combustion of the dried embryos in a sample

oxidizer. Perfused and intact seed coats were extracted in hot (50 0 C) 80% (v Iv)

ethanol and the extracts passed through tandem cation and anion exchange

columns to produce a neutral fraction (primarily sugars). The 14C content of

neutral column eluates was determined by liquid scintillation counting and the

eluate remaining after counting was analyzed for sucrose + glucose by

sucrose inversion followed by the glucose oxidase/peroxidase procedure

(Berg meyer and Bernt, 1974). The 14C remaining in ethanol-extracted seed

coats was quantified by liquid scintillation counting. Aliquots of 10 minute

samples of perfusate were liquid scintillation counted, and the remainder was

assayed for sucrose + glucose as described above for neutral column eluates.

The specific activity of sucrose + glucose in samples of perfusate or

neutral column eluate was calculated based on 12 moles of carbon atoms per

mole of sucrose + glucose assayed, by assuming that 100% of the 14C in the

sample was sucrose + glucose, and that glucose and fructose were present at

equimolar concentrations. If the sample contained a significant quantity of 14C

137

in a substrate besides sucrose or glucose, the calculated sucrose + glucose

specific activity would be over-estimated, while a decrease in fructose relative

to glucose in the sample would lead to underestimation of the sucrose +

glucose specific activity.

Normalization of tracer data. The specific activity of 14C02 used for

labelling was used to normalize the tracer data of experiments which used

different 14C02 specific activities, so that comparisons between experiments

could be made. This was accomplished by multiplying all 14C and specific

activity measurements within an experiment by a dimensionless quantity

derived by inverting the specific activity of 14C02 used in the experiment, and

multiplying by the units of specific activity (G8q (mol Cr1). Thus, the highest

specific activity that could be reached should be 1.0 G8q mol-1 C; higher

specific activities were the result of the presence of 14C in compounds other

than sucrose, glucose, and fructose.

Size adjustment for perfused seeds. To facilitate comparisons

between intact and perfused seeds, a seed size correction factor based on the

sugar content of the basal portion of cut seed coats (approximately 70% of

intact seed coat sugar content) was used to adjust perfused seed data for the

fraction of the seed that was lost due to excision (Chapter 3).

RESULTS

Rapid change effects on photosynthate efflux. The rapid alterations

of sugar (sucrose + glucose) and 14C efflux that were induced by changes in

the mannitol and EDTA concentration of the perfusion solution were often quite

variable between experiments. The extent of this variability depended on the

138

treatment and is illustrated by changes in the size of the standard error bars of

the data presented in Figures 4.1 to 4.4. In order to simplify the description of

data presented in this section, the effects of treatments will be described in

relation to the time at which the treatments were applied.

After the initial 4 hours of perfusion, a change in the mannitol

concentration of the perfusion solution from 100 to 10 mM or from 10 to

100 mM did not alter the efflux of 14C or sugar from the seed coat (Figs 4.1

and 4.2). In addition, the rate of increase in the specific activity of eluted sugar

remained relatively constant after changing the mannitol concentration from

100 to 10 mM or from 10 to 100 mM (Figs 4.1 and 4.2). Changing the mannitol

concentration in the perfusion solution from 500 to 10 mM or from 500 to

100 mM produced a rapid increase in the efflux of both 14C and sugar, but the

transition from 500 to 10 mM mannitol produced a greater increase in efflux

than did the change from 500 to 100 mM (Figs 4.3 and 4.4). The specific

activity of sugar released from the seed coat following the transition from

500 mM to 10 or 100 mM mannitol did not appear to increase, as it had during

the initial 4 hours of perfusion (Figs 4.3 and 4.4).

Following 6.5 hours of perfUSion, a change in mannitol concentration

from 10 to 500 mM produced a negligible increase in sugar and 14C efflux. In

all three replications, the change from a 10 to a SOO mM mannitol concentration

in the perfusion solution led to a relatively constant eluted sugar specific activity

over time, although, in one replication out of three, there was an initial decrease

in the specific activity of eluted sugar caused by an increase in sugar efflux that

was larger than the increase in 14C efflux (Fig. 4.2). A change in mannitol

concentration from 100 to 500 mM produced a small increase in sugar and

14C efflux, and the specific activity of eluted sugar increased at approximately

139

Figure 4.1 Time course of seed coat perfusion with changing mannitol

concentrations with or without 15 mM EDTA. The concentration of mannitol

and the presence of EDT A in the perfusion solution is indicated at the top of the

figure. A, sucrose + glucose efflux; 8, 14C efflux; C, specific activity of sucrose

+ glucose. Data are the mean ± SE of three experiments. Tracer data were

normalized to a 14C02 specific activity of 1.0 GBq mol-1 (see Materials and

Methods). Zero time refers to beginning of seed coat perfusion.

~

I ..c o E :t

2

1

10 mM 100 mM 500 mM' 500 mM+EDTA

A-Suc+Glc Efflux

J. ...... 0 ~ 6~ ~ 00 bOoocPr;:dJd.. ~~TII 0 ~l,~ __ .L .L _ •• _ """"'"2f 0

1 5 ~ 8-14C Efflux l:ql~

I!~J. ... I.e 10 C"

OJ ~ 5

I 0.8 ~ 0.6 CT 0.4

OJ (!) 0.2

o

AAi! ~ AJ. ... UA6A IIX A: -J.M A ~!"h_.'~u~Qi~iA -.~ TTTI~-~ ,~_ -I -"'J.! ~J.

.L~ ~J.J.J.

C-Suc+Glc S.A. ~~ OL

2 4 6 8 10 12 Time (hours)

..a. ~ o

141

Figure 4.2 Time course of seed coat perfusion with changing mannitol

concentrations with or without 15 mM EDT A. The concentration of mannitol

and the presence of EDTA in the perfusion solution is indicated at the top of the

figure. A, sucrose + glucose efflux; B, 14C efflux; C, specific activity of sucrose

+ glucose. Data are the mean ± SE of three experiments. Tracer data were

normalized to a 14C02 specific activity of 1.0 GBq mor1 (see Materials and

Methods). Zero time refers to beginning of seed coat perfUSion.

142

N ~

« r-0 w + ~ a E ~

0 0 ~

~ E

0 0 LO ~

en '-:::s 0

co..c: '-"

~ Q) E E 0 .-r-

X ::J

~ ;;: .... E lLJ x

0 ::J

0 ;;: (!) ....

0 + lLJ

0 (.)

N

:::J ~ (/l -I I « o~---1 m

a N ~ LO 0 U') OOCO~N - ~

. . . . o 0 0 0

L_4 lOW". ~_4 bS>t ~_IOW b8~

143

Figure 4.3 Time course of seed coat perfusion with changing mannitol

concentrations with or without 15 mM EDT A. The concentration of mannitol

and the presence of EDTA in the perfusion solution is indicated at the top of the

figure. A, sucrose + glucose efflux; B, 14C efflux; C, specific activity of sucrose

+ glucose. Data are the mean ± SE of three experiments. Tracer data were

normalized to a 14C02 specific activity of 1.0 GBq mol-1 (see Materials and

Methods). Zero time refers to beginning of seed coat perfusion.

144

N ..-

« I-0 W

+ ::2! t-<I-t 0 E ~~

..-

0 t--<I---t

« -I I- -I

~-t 0 ~~ w

+ a~ ::2! E ~-t

0 ~-t ".......

I-<:I-t 0 ~' en

LO L.. ::J

t-<I .. 0 t-<I~ (0-'=

~:!t '-' ::2! ~-t Q)

E ,.!-3~ E 0 .-

" <J~ t-.....-<2--1

<J

X :J

~ ::2! ;;:: "'"" E I.&J X Ul

:J 0 ;;:: 0

"'"" (!) 0 I.&J + LO U 0

~ :J .... Ul I I m u

0 N ..- to V N V n N ..-. . . .

0 0 0 0

l_4lown l-4 bS>! l_IOW bSD

145

Figure 4.4 Time course of seed coat perfusion with changing mannitol

concentrations with or without 15 mM EDT A. The concentration of mannitol

and the presence of EDTA in the perfusion solution is indicated at the top of the

figure. A, sucrose + glucose efflux; 8, 14C efflux; C, specific activity of sucrose

+ glucose. Data are the mean ± SE of four experiments. Tracer data were

normalized to a 14C02 specific activity of 1.0 GBq mol-1 (see Materials and

Methods). Zero time refers to beginning of seed coat perfusion.

146

N ---... ~

« ----t

..... ~.-!. .. 0 .~~

t-<I-i W 1 1 44--!.

+ ~--. ~ I. 11~ 0 E ..-.c3---1

1 ~---11 ~

0 0 .--<J--t to ~-1 « f ~~ ..... f

0 ] 11 w CD +

~ E ~~ 1-4-t

1. 0 o~ ,-... 0 ~ (I)

La

1 ~-t ::l 0

J (0£

:?; ~

E

l 1<41 Q)

t<tl E 0 t<tt ~~

.-0 I-

~ 1-<1"

X 1 1: ::J 1 \;: f ~ ~ ~ 1

E LaJ t X (I) .. ~: o o~ ::J N 0 - 0g-t \;: I~ " ~

~

0 +0 ...... LaJ 1, to o "']--t CJ <I

::J~-t V (I) --t .... ---1 l--ot ~ -t

0 t") N ~ 0 IX) tOvN to V N

..- . . . 0 0 0

~_4 lown ~_4 b8>! ~_IOW b8~

147

the same rate as it did before the change in mannitol concentration (Fig. 4.1).

The effect of changing from 100 mM mannitol to 100 mM mannitol + EDTA

was a minimal increase in sugar and 14C efflux, and there was a also a slow

increase in the specific activity of sugar eluted following EDTA inclusion in the

perfusion solution (Fig. 4.4). After a change from a 10 mM mannitol solution to

one containing both 500 mM mannitol and EDT A there was a small increase in

the efflux of sugar and 14C, and the specific activity of eluted sugar appeared

to increase gradually (Fig. 4.3).

After 9 hours of perfUSion, a change from 500 mM mannitol to 500 mM

mannitol + EDTA in the perfusion solution caused a slow but steady increase

in unlabeled sugar and 14C efflux, and the specific activity of sugar eluted

following the change increased at a constant rate (Fig. 4.1). The change from a

500 mM mannitol solution to a solution containing both 10 mM mannitol and

EDTA produced a very large and rapid increase in both unlabeled sugar and

14C efflux (Fig. 4.2). The specific activity of eluted sugar appeared to increase

abruptly at the transition from 500 mM mannitol to 10 mM mannitol + EDTA

(Fig. 4.2), and this effect was consistent between replications (individual

replications not shown). Following the initial increase, the specific activity of

eluted sugar did not appear to increase greatly over time (Fig. 4.2). There was

a small increase in the efflux of unlabeled sugar and 14C when the mannitol

concentration of a solution containing EDTA was changed from 500 to 10 mM

or from 100 to 500 mM, with the change from 100 to 500 mM exhibiting the

greatest increase in efflux (Figs 4.3 and 4.4). The specific activity of sugar

eluted into solutions containing EDTA increased following the transition from

100 to 500 mM mannitol (Fig. 4.4), while the eluted sugar specific activity

appeared to level off following the change from 500 to 10 mM mannitol (Fig.

4.3).

148

Cumulative effects of mannitol andEDTA treatments on

photosynthate retention and release from seed coats. Of the four

experiments described in Figures 4.1 to 4.4, the treatments illustrated in Figure

4.1 were associated with the highest quantity of sugar and 14C released from

and retained in seed coats following perfusion (Table 4.1). In addition, the

specific activity of sugar eluted from and remaining in the seed coats treated as

illustrated in Figure 4.1 was higher than for all the other experiments in this

study (Fig. 4.1, Table 4.1). The cumulative effect of the treatments illustrated in

Figure 4.2 was the second highest specific activity of sugar eluted from and

remaining in the seed coat, and the second highest quantity of 14C eluted from

and remaining in the seed coat. In contrast, this series of treatments (Fig. 4.2)

was also associated with the lowest quantity of unlabeled sugar eluted from

and remaining in seed coats that was observed for any of the experiments in

this study (Table 4.1). The treatments illustrated in Figures 4.1 and 4.2 were

associated with greater imports of 14C to seed coats than were observed in

experiments with constant mannitol concentrations in the perfusion solution

(Chapter 3; Table 4.1). The specific activity of sugar remaining in seed coats

perfused as described in Figures 4.1 and 4.2 was in some cases higher than

the specific activity of sugar extracted from intact seed coats, although this

condition was highly variable for experiment 4.1 (Table 4.1). This may be

explained by the presence of uncharged, 80% ethanol soluble compounds in

the perfused seed coats, but the nature of these compounds is unknown and

the high variability in these measurements may indicate some degree of

experimental error.

Data for the cumulative amounts of 14C and unlabeled sugar eluted

from and remaining in seed coats treated as illustrated in Figures 4.1 to 4.4 is

TABLE 4.1 The cumulative effects of mannitol and EDTA treatments on the quantity of sugar (sucrose + glucose) and

14C eluted from and remaining in perfused seed coats. The four rows of this table represent data from the experiments

illustrated in Figures 4.1 to 4.4. Data for the sugar and 14C eluted from seed coats during perfusion (Eluted), remaining

in the seed coats after perfusion (Seed coat), and the combined total of the amount eluted from and remaining in the

seed coats after perfusion (Total) are expressed as a percentage of the quantity present in intact seed coats. The ratio

of the amount of sugar or 14C eluted from the seed coat to the quantity retained in the seed coat following perfusion is

presented in the columns titled "Ratio". The total amount of 14C imported to perfused seed coats was expressed as a

percentage of the total amount of 14C imported to intact seeds (seed coat + embryo + cotyledons) in the column titled

"Total 14C as percent of intact seed". The specific activity of sugar remaining in seed coats following perfusion is ex­

pressed as a percentage of the specific activity of sugar in intact seed coats in the final column of this table. Each data

point is the mean of three experiments, and the mean.±. SE is presented for selected data. Tracer data were normalized

to a 14C02 specific activity of 1.0 GBq mol-1 (see Materials and Methods)

14C Total 14C Seed coat

Sugar as percent sugar of intact specific

Figure Seed coat Eluted Ratio Total Seed coat Eluted Ratio Total seed activity number (%) (%) (E/S) (%) (%) (%) (E/S) (%) (%) (%)

4.1 101.5.±. 32.9 127.7 .±. 1.4 1.26 229.2 104.0.± 26.6 63.3.±. 4.8 0.61 167.3 59.4 139.3 .±. 32.3

4.2 59.2.±. 7.0 76.4 ±. 12.9 1.29 135.6 70.6.±24.7 47.0 ±. 5.5 0.66 117.6 37.6 95.6.± 7.1

4.3 65.2.±. 21.0 118.6 .±. 38.8 1.80 183.8 22.7 .± 4.8 22.4 .±. 6.0 0.99 45.1 15.1 42.3.±. 7.1

4.4 81.6.±. 8.3 106.6.±. 12.5 1.30 188.2 41.8.± 4.7 34.9.±. 7.1 0.83 76.7 27.3 60.6.±. 10.1

~

~

150

summarized in Table 4.1. The long term effects of the treatments illustrated in

Figures 4.1 to 4.4 on the ratio of unlabeled sugar and 14C eluted to that

remaining in the seed coat is also presented in Table 4.1.

DISCUSSION

Long term changes In phloem Import and photosynthate

partitioning in seed coats. The compartmentation of unlabeled sugar and

14C (primarily 14C-labeled sugar) within seed coats probably changes during

long term perfusion periods such as the 12 hour period used for these

experiments (Thorne and Gifford, 1986; Minchin and Grusak, 1988; Minchin

and Thorpe, 1989; Chapter 3). In addition, the application of various treatments

during the perfusion period may cause either temporary or long term

alterations in the compartmentation of 14C and sugar within the seed coat.

Therefore, analysis of the cumulative effects of a series of treatments given

during the perfusion period is complex, and the effects of a specific treatment

cannot be determined based on the cumulative results of a series of

treatments. In some cases, however, a specific series of treatments produced

striking changes in the net import of sugar and 14C to perfused seed coats, or

significantly altered the relative partitioning of photosynthate between the seed

coat and the perfusion solution. For this reason, associations could be made

between the cumulative results of a series of treatments and the general

properties of this series of treatments, such as the length of EDTA treatment, or

the period and duration of perfusion with specific osmoticum concentrations. It

must be recognized that these associations are speculative in nature and may

provide only circumstantial evidence for the effects of various treatments.

151

When all four treatment series in this study were compared (Figs 4.1 to

4.4), it was found that the treatment series which had the highest net imports of

14C also had the highest specific activities of sugar eluted from and remaining

in the seed coats after perfusion (Table 4.1). This observation conflicts with the

results presented for experiments with constant mannitol concentrations,

because high 14C imports were associated with low specific activities of sugar

eluted from and remaining in the seed coat after perfusion (Chapter 3). The

association of low sugar specific activities with high 14C import rates was

explained by the hypothesis that high mannitol concentrations, which were

found to increase the net import of 14C, also favored the retention of sugar in

the seed coat which thus reduced the increase in seed coat sugar specific

activity (Chapter 3). In the current study, the association of high 14C imports

with high sugar specific activities may be due to the introduction of low

concentrations of osmoticum before high concentrations at some point during

every treatment series of this study. Treatment with a low osmoticum

concentration would be expected to enhance the net loss of sugar from the

seed coat (Patrick, 1984; Patrick et aI., 1986; Wolswinkel et aI., 1986; Grusak

and Minchin, 1988; Chapter 3), while treatment with a high osmoticum

concentration would be expected to increase phloem import and the net

accumulation of sugar in the seed coat (Minchin and McNaughton, 1986;

Wolswinkel et aI., 1986; Grusak and Minchin, 1988; Minchin and Grusak, 1988;

Chapter 3). Thus, the sequential treatment of seed coats with a low followed by

a high concentration of osmoticum may have led to the net loss of sugar from

the seed coat, followed by the replacement of lost sugar with sugar imported

by the phloem. The replacement of the low specific activity sugar originally

152

present in the seed coat with the high specific activity sugar imported by the

phloem would explain the dependence of seed coat sugar specific activity on

the net import of 14C by the phloem.

In contrast with the treatment series illustrated in Figures 4.1 and 4.2, a

high concentration of osmoticum was present during the first 4 hours of the

perfusion period in the treatment series illustrated in Figures 4.3 and 4.4.

Phloem import was probably stimulated by high osmoticum concentrations in

the perfusion solution (Minchin and McNaughton, 1986; Wolswinkel et aI.,

1986; Grusak and Minchin, 1988; Chapter 3), and, because the specific activity

of phloem sugar was lowest during the early hours of the perfusion period,

phloem import stimulation during the initial hours of perfusion by a high

concentration of osmoticum in the perfusion solution would be expected to

enhance the import of low specific activity sugar. Thus, perfusion with a high

concentration of osmoticum during the first 4 hours of the perfusion period

may have produced the high cumulative import of unlabeled sugar and the low

cumulative import of 14C that was associated with the treatment series

illustrated in Figures 4.3 and 4.4 (Table 4.1). In addition, the enhanced import

of low specific activity sugar at the beginning of the perfusion period may

explain the relatively low specific activity of sugar eluted from and remaining in

the seed coats treated as illustrated in Figures 4.3 and 4.4 (Table 4.1).

The net import of 14C and sugar to the seed coats treated as illustrated

in Figure 4.1 was greater than for any other series of treatments in this chapter,

and was also greater than observed in experiments with constant osmoticum

concentrations in the perfusion solution (Chapter 3; Table 4.1). This could have

been due to a high mannitol concentration (500 mM) in the perfusion solution

during the final 5.5 hours of perfusion, but the resulting stimulation of import

153

was greater than that observed when 500 mM mannitol was present over the

entire 12 hour perfusion period (Chapter 3; Table 4.1). Another possibility is

that combined treatment with 500 mM mannitol and EDTA was responsible for

the high sugar and 14C import associated with the treatments illustrated in

Figure 4.1, but the same treatment (500 mM mannitol + EDTA) was also

applied in the experiment described in Figure 4.3, which had the lowest 14C

import of all the experiments. The total amount of 14C and sugar eluted from

the seed coat was relatively low compared with the amount present in the seed

coat at the end of the perfusion period in the experiments illustrated in Figure

4.1, and this may have been due to an increase in the retention of unlabeled

sugar and 14C in the seed coat caused by treatment with high osmoticum

concentrations near the end of the perfusion period. The high sugar and 14C

import associated with the treatments illustrated in Figure 4.1 may be the result

of the differential stimulation of net sugar release and net phloem import at

different times during the perfusion period, but this explanation remains purely

hypothetical.

The treatments illustrated in Figure 4.2 induced the second highest 14C

import to seed coats that was observed in these experiments, but this series of

treatments also produced the lowest apparent import of unlabeled sugar

(Table 4.1). The treatments illustrated in Figure 4.2 were similar to those

illustrated in Figure 4.1, with the exception that a low osmoticum concentration

was applied near the end of the perfusion period in Figure 4.2, while a high

osmoticum concentration was applied during the same period in Figure 4.1.

The transition from a high to a low osmoticum concentration + EDTA at the

end of the treatment series in Figure 4.2 induced a very large and rapid release

of relatively high specific activity sugar. The release of relatively high specific

154

activity sugar at the end of the perfusion period may have represented the

specific stimulation of photosynthate efflux from the phloem, and this would

explain the high 14C import observed for this treatment series, while the

relatively low sugar import observed for this treatment series may be the result

of the small amount of time that high osmoticum concentrations were present

in the perfusion solution. The high specific activity of sugar remaining in the

seed coats treated as illustrated in Figure 4.2 may be explained by sequential

treatment with a low followed by a high osmoticum concentration in the

perfusion solution as described above for the treatment series illustrated in

Figure 4.1.

The application of EDTA during the final 5.5 hours of perfusion (Figures

4.3 and 4.4) did not lead to a greater net import of sugar or 14C than when

EDTA was applied solely for the final 3 hours of perfusion (Figures 4.3 and 4.4)

(see Table 4.1). In fact, longer term treatment with EDTA (5.5 hours versus 3

hours) was associated with lower 14C import, and an increase in the release of

unlabeled sugar and 14C relative to the amount remaining in the seed coat

(Table 4.1). Thus, the results presented above may indicate that longer EDTA

treatments did not enhance phloem import, but did increase the release of

photosynthate from the seed coat. These results support the hypothesis that

EDTA may produce a general leakiness in seed coat membranes leading to a

net loss of photosynthate from the seed coat. The difference in the length of

time EDTA was presented was not great in the comparisons described above

(5.5 versus 3 hours) and this may reduce the resolution of EDTA effects on

photosynthate unloading. In addition, the interaction of EDTA with specific

osmoticum treatments may have been greater than the effects of EDTA alone,

and thus the results presented above are not clear evidence for the mechanism

by which EDTA alters photosynthate unloading from seed coats.

155

The specific activity of eluted sugar was overestimated in these

experiments (and in those of Chapter 3) because sugar was not purified before

specific activity determinations were made (see Materials and Methods). For

this reason, eluted sugar specific activity measurements were not equivalent to

seed coat sugar specific activity measurements. However, the specific activity

of sugar eluted at the end of the perfusion period was commensurate with the

specific activity of sugar remaining in the seed coat at the end of the perfusion

period in all of the experiments in this study and in Chapter 3 (Figures 4.1 to

4.4). The relatively close relationship between the specific activity of sugar

eluted near the end of the perfusion period and the specific activity of sugar in

the seed coat may indicate that unlabeled and 14C-labeled sugar were

uniformly distributed within the seed coat near the end of the perfusion period.

Apoplastic sugar exchange between the phloem and other cells in the seed

coat was probably inhibited by the rapid removal of apoplastic solutes by

perfusion solutions (Chapter 3). Thus, the uniform distribution of unlabeled and

14C-labeled sugar within the seed coat is consistent with the hypothesis that

symplastic photosynthate movement between the phloem and other cells of

the seed coat is the primary route of phloem unloading.

Osmoticum-induced changes in photosynthate efflux. The

observation that rapid changes in osmoticum concentration produce

immediate changes (within minutes) in sugar and 14C efflux from seed coats

demonstrates that the rapid modification of cell turgor may induce changes in

photosynthate unloading from seed coats (Patrick, 1984; Patrick et aI., 1986;

Minchin and McNaughton, 1986; Ellis and Spanswick, 1987; Grusak and

Minchin, 1988). Small changes in osmoticum concentration (from 10 to

100 mM mannitol or vice versa) did not elicit significant changes in sugar or

156

14C efflux, while significant changes in efflux were usually observed in

response to larger changes in osmoticum concentration (from 10 or 100 mM to

500 mM mannitol, or vice versa). After 4 hours of perfusion, the magnitude of

the increase in sugar and 14C efflux caused by an osmoticum concentration

change appeared to be relative to the magnitude of the change in osmoticum

concentration. For example, the increase in efflux produced by a change from

500 to 10 mM was greater than the increase for a change from 500 to 100 mM

mannitol, while no observable change in efflux was observed following a

change in osmoticum concentration from 100 to 10 mM or from 100 to 10 mM.

The lack of an increase in unlabeled sugar and 14C efflux when the osmoticum

concentration in the perfusion solution was changed from 100 to 10 mM

contrasts with results obtained using Phaseolus vulgaris (Patrick, 1984; Patrick

et aI., 1986), and soybean (Ellis and Spanswick, 1987). The reason for the

difference between this study and previous studies is unknown, although the

relatively low time resolution of the current study may have been at least

partially responsible for this discrepancy.

The increase in unlabeled sugar and 14C efflux in response to

osmoticum concentration changes was not always comparable to the relative

change in osmoticum concentration. This was demonstrated by the contrast

between the negligible change in sugar and 14C efflux induced by a mannitol

concentration change from 10 to 500 mM after 6.5 hours of perfusion (Fig. 4.2),

while the change from 100 to 500 mM elicited a significant increase in sugar

and 14C release (Fig. 4.1). Ellis and Spanswick (1987) demonstrated that an

osmoticum concentration change from high to low produced a rapid peak in

sugar efflux, while the response to an osmoticum concentration change from

low to high was a large increase in sugar efflux followed by a relatively slow

157

decline. The two different types of efflux stimulation were said to be caused by

the differential activation of two distinct mechanisms of efflux regulation, one of

which was stimulated by low cell turgor and one which was stimulated by high

cell turgor (Ellis and Spanswick, 1987). In this study, the lack of clear

differences in the characteristics of efflux in response to different types of

osmoticum concentration changes may indicate that changes in efflux were

simply the result of any significant change in osmoticum concentration.

However, differences in efflux kinetics were sometimes obscured in the current

study by the effect of averaging the results of multiple repetitions of the same

experiment (individual experiments were presented by Ellis and Spanswick,

1987), because changes in efflux were not precisely synchronized in time

between different repetitions. As was previously demonstrated by Ellis and

Spanswick (1987), stimulations of sugar (and 14C) efflux could be induced by

changing the osmoticum concentration from high to low or by changing the

osmoticum concentration from low to high, and this may support the

hypothesis that more than one turgor-sensitive mechanism may regulate

photosynthate efflux from seed coats.

Osmoticum-induced changes in the specific activity of eluted

sugar. When seed coats were perfused with constant concentrations of

osmoticum, the specific activity of eluted sugar typically increased over time

throughout the 12 hour perfusion period, although there was some evidence

for a lag in specific activity increase between 7 and 9 hours after the initiation of

perfusion (Chapter 3). Osmoticum concentration change treatments applied

6.5 hours after the initiation of perfusion produced ambiguous results because

the rate of increase in eluted sugar specific activity may have declined

independently of treatments 7 to 9 hours after the initiation of perfusion. After

158

4 hours of perfusion, however, some osmoticum concentration change

treatments appeared to halt the increase in eluted sugar specific activity,

although treatments that enhanced the rate of increase in eluted sugar specific

activity were poorly resolved due to the lack of adequate controls.

Four hours after the initiation of perfusion, small changes in osmoticum

concentration (10 to 100 mM or 100 to 10 mM) did not significantly alter the

rate of increase in eluted sugar specific activity. However, changing from a high

osmoticum concentration (500 mM) to a lower osmoticum concentration (10 or

100 mM) appeared to inhibit the increase in eluted sugar specific activity.

Phloem import is probably reduced by low concentrations of osmoticum in the

perfusion solution (Minchin and McNaughton, 1986, Wolswinkel et aI., 1986;

Grusak and Minchin, 1988; Chapter 3), and thus the increase in eluted sugar

specific activity may have been inhibited by a sudden decrease in the rate of

phloem import of high specific activity sugar caused by lowering the

osmoticum concentration in the perfusion solution. The observation that sugar

and 14C efflux were stimulated by the transition from a high to a low

osmoticum concentration, while the specific activity of eluted sugar stopped

increasing, is consistent with a stimulation of sugar release from cells within the

seed coat that were not as highly labeled as the phloem. This may support the

hypothesis that lower osmoticum concentrations specifically stimulate the

release of sugar from cells near the inner surface of the seed coat (Patrick,

1984; Patrick et aI., 1986).

In general, the specific activity of sugar eluted from seed coats did not

change dramatically following a change in osmoticum concentration. The only

exception to this generality was the apparent decrease in eluted sugar specific

activity induced by a change from 10 to 500 mM mannitol in the perfusion

159

solution (Fig. 4.2), and this was not consistently observed (present in only one

out of three replications). The general lack of rapid changes in the specific

activity of eluted sugar may demonstrate that labeled and unlabeled sugar

were fairly well mixed within the seed coat. This implies that the bulk of seed

coat sugar exchanges to some degree with the high specific activity sugar

imported by the phloem, perhaps due to the symplastic movement of sugars

from the phloem throughout the seed coat before efflux to the apoplast.

Another possibility is that osmoticum concentration change treatments were

not capable of specifically stimulating sugar effluxes from individual sugar pools

within the seed coat that may have had different sugar specific activities. In

either case, the observation that rapid stimulations of sugar and 14C efflux did

not lead to rapid changes in the specific activity of eluted sugar may indicate

that osmoticum concentration change treatments did not severely disrupt the

compartmentation of sugars within the seed coat.

Osmoticum and EDTA-induced changes in photosynthate

unloading. When the perfusion solution was changed from a solution without

EDTA to a solution with EDTA, the efflux of sugar and 14C was increased in all

experiments. The magnitude of EDT A-induced increases in sugar and 14C

efflux appeared to depend on the osmoticum concentration in the perfusion

solution both before and after EDTA treatment. For example, the change from

a 500 mM mannitol solution to a solution containing both 10 mM mannitol and

EDTA produced a very large and rapid increase in both sugar and 14C efflux

(Fig. 4.2), while the effect of a change from 100 mM mannitol to 100 mM

mannitol + EDTA was a minimal increase in sugar and 14C efflux (Fig. 4.4).

The magnitude of sugar and 14C efflux following EDTA treatment was not

relative to the size of the change in osmoticum concentration that occurred

160

simultaneously with EDTA treatment. This was demonstrated by the large

increase in unlabeled sugar and 14C efflux caused by the change from a

500 mM mannitol solution to a solution containing both 10 mM mannitol and

EDTA, while the change from a 10 mM mannitol solution to one containing both

500 mM mannitol and EDTA produced a small increase in the efflux of sugar

and 14C (Fig. 4.3).

The specific activity of eluted sugar typically increased at a fairly

constant rate following EDTA treatments, except when a low concentration of

mannitol (10 mM) was present in the perfusion solution. The observation that

the specific activity of eluted sugar continued to increase following most EDTA

treatments indicates that phloem import to the seed coat was not inhibited by

EDTA treatment. The observation that the specific activity of eluted sugar did

not increase over time following treatment with EDTA and a low concentration

of osmoticum may be explained by a reduction in the phloem import of high

specific activity sugar to the seed coat by treatment with a low concentration of

osmoticum (Minchin and McNaughton, 1986, Wolswinkel et aI., 1986; Grusak

and Minchin, 1988; Chapter 3).

EDTA treatments sometimes produced changes in photosynthate efflux

that were quite different from the effects of osmoticum concentration change

treatments in these experiments, or in the experiments of Ellis and Spanswick

(1987) on soybeans. For example, unlabeled sugar efflux was stimulated to

such a great extent by the change from a 500 mM mannitol solution to a

solution containing both 10 mM mannitol and EDTA, that efflux was enhanced

above the rate present during the first hour of perfusion (Fig. 4.2). In addition,

the efflux of 14C stimulated by the above treatment was greater than any other

observed in these experiments. There was a rapid decline in efflux following the

161

initial increase, and this may have been due to the net loss of sugar from the

seed coat, because a relatively low level of sugar remained in the seed coat

following this treatment (Table 4.1). When EDTA was present in the perfusion

solution both before and after the change from 500 to 10 mM mannitol (Fig.

4.3), no large increase in sugar and 14C efflux was observed, which indicates

that the very large efflux observed in Figure 4.2 was due to the introduction of

EDTA to the perfusion solution at the same time as the osmoticum

concentration was changed. The induction of a high degree of membrane

leakiness by treatment with EDTA would be expected to prevent seed coats

from responding to changes in cell turgor, and this may explain the reduction in

the effect of an osmoticum concentration change from 500 to 10 mM when

EDTA was present in the perfusion solution previous to the osmoticum

concentration change.

The specific activity of eluted sugar increased rapidly following the

change from 500 mM mannitol to 10 mM mannitol + EDTA in all three

replications of this experiment, but, following the initial increase, the specific

activity of eluted sugar did not appear to increase over time (Fig. 4.2). The

rapid increase in the specific activity of eluted sugar may have been due to the

direct stimulation of sugar efflux from the phloem, which would have been at a

relatively high specific activity. However, the lack of an increase in eluted sugar

specific activity following the initial stimulation may indicate that phloem import

was inhibited by this treatment, perhaps due to a low concentration of

osmoticum in the apoplast. The large increase in sugar and 14C efflux that was

described above remains enigmatic, although it appears that there may have

been a direct stimulation of photosynthate efflux from the seed coat phloem,

perhaps without a significant increase in phloem import to the seed coat.

162

Another unusual effect of EDTA treatment was the slow increase in

unlabeled sugar and 14C efflux observed following the addition of EDTA to the

perfusion solution when the osmoticum concentration remained at 500 mM

mannitol (Fig. 4.1). This was the only example of a slow but large increase in

unlabeled sugar and 14C efflux in these experiments, and in those of Ellis and

Spanswick (1987), and was comparable to the stimulation of sugar efflux from

Vicia faba seed coats that was caused by treatment with 400 mM mannitol +

15 mM EDTA (Wolswinkel, 1987a). The specific activity of eluted sugar

continued to increase following the addition of EDTA to the perfusion solution,

and this may indicate that phloem import of high specific activity sugar

continued following EDTA treatment. In addition, the very high specific activity

of sugar eluted following EDTA treatment may indicate that the release of sugar

from the phloem may have been specifically enhanced. The exact nature of

EDT A-induced stimulations of sugar and 14C efflux remains speculative, but,

high osmoticum concentrations probably enhance net phloem import (Minchin

and McNaughton, 1986, Wolswinkel et aI., 1986; Grusak and Minchin, 1988,

Chapter 3), and thus EDTA may have only stimulated the non-specific release

of photosynthate from the seed coat.

Cellular damage due to rapid changes in osmoticum concentration.

Large and rapid changes in osmoticum concentration may have caused the

plasmolysis of seed coat cells and thus damaged the plasmodesmatal

connections between cells in the seed coat. Although this possiblity was not

evaluated microscopically, it should be noted that large changes in osmoticum

concentration (500 to 100 mM or 500 to 10 mM) near the beginning of the

perfusion period were associated with low seed coat and perfusate sugar

163

specific activities (Figures 4.3 and 4.4; Table 4.1). This may have been the

result of the disruption of symplastic continuity within the seed coat because

the cells that were isolated from the phloem by plasmodesmatal damage could

not accumulate the high specific activity sugar imported by the phloem.

CONCLUSIONS

Changes in osmoticum concentration from high (500 mM) to low (10 or

100 mM), or from low to high, both stimulated the efflux of unlabeled sugar and

14C. However, the specific activity of eluted sugar increased following the

change from a high to low concentration of osmoticum in the perfusion

solution, while the change from a low to a high osmoticum concentration

appeared to halt the increase in eluted sugar specific activity. The results of

these experiments are consistent with the inhibition of phloem import by high

cell turgor in the seed coat (Minchin and McNaughton, 1986; Wolswinkel et aI.,

1986; Grusak and Minchin, 1988; Chapter 3), and the stimulation of a turgor­

sensitive efflux control mechanism by high cell turgor (Patrick, 1984; Patrick et

aI., 1986; Wolswinkel et aI., 1986; Grusak and Minchin, 1988; Chapter 3). In

general, rapid changes in the specific activity of eluted sugar did not occur,

which may demonstrate that labeled and unlabeled sugar were fairly well mixed

within the seed coat, or that osmoticum concentration change treatments did

not stimulate efflux from specific compartments within the seed coat. The

specific activity of sugar eluted near the end of the perfusion period paralleled

the specific activity of sugar in the seed coat, and this may provide further

evidence that unlabeled and 14C-labeled sugar were uniformly distributed

within the seed coat. The hypothesis that sugar imported by the phloem moves

symplastically throughout the seed coat before efflux to the apoplast (Patrick

164

and McDonald, 1980; Offler and Patrick, 1984; Patrick et aI., 1986; Grusak and

Minchin, 1988; Chapter 3) was supported by results which indicated that

unlabeled and 14C-labeled sugar were uniformly distributed within perfused

seed coats, because apoplastic solute movement from the phloem to other

cells was probably inhibited by seed coat perfusion.

EDTA treatments always enhanced the efflux of unlabeled sugar and

14C from seed coats, although the magnitude of EDTA-induced increases in

efflux appeared to depend on the osmoticum concentration in the perfusion

solution, both before and after EDTA treatment. In some experiments, the

interaction of EDTA with specific osmoticum treatments may have been greater

than the additive effects of EDTA or osmoticum change treatments alone. In

addition, the increases in unlabeled sugar and 14C efflux that were associated

with EDTA treatments were in some cases much larger or had different kinetics

than any that were attributed to osmoticum concentration change treatments in

this study. The effects of EDTA treatments were complex and may have

involved general increases in efflux from the seed coat, as there was no reliable

evidence for the direct stimulation of phloem import by EDT A. Although the

exact nature of EDT A-induced stimulations of sugar and 14C efflux remain

speculative, it appears that under some conditions EDTA may be capable of

direct stimulation of photosynthate efflux from the phloem.

CHAPTERS

Concluding Remarks and Directions for Future Research

Improvement of Techniques

Steady state labelling. The methods used for steady state labelling in

this study were limited by the low C02 concentrations present in the leaf

chamber during the labelling period (approximately 200 J.'L L-1), and the lack of

direct monitoring and control over the specific activity of 14C02. The low C02

concentrations present in the leaf chamber were the result of the very large

leaves that were labeled in these experiments. The total photosynthetic area

was large and thus the net photosynthetic rate of the leaf was greater than the

supply of C02 to the leaf chamber. The flow rate of C02-containing air into the

leaf chamber was relatively high during these experiments (approximately

3.7 L min-1), but the concentration of C02 in the air entering the leaf chamber

was maintained at a fairly moderate level (350 J.'L L-1). The C02 concentration

of air entering the leaf chamber could be increased in future experiments

without any significant changes in experimental protocol, as C02

concentrations were under software control and could be raised to a level

commensurate with the photosynthetic rate of the leaf under study. The

specific activity of 14C02 was probably fairly constant during the labelling

period, because all of the 14C02 added to the system was from a source with

a constant 14C02 specific activity. However, there was some evidence that the

specific activity of 14C02 was relatively low during the early hours of the

labelling period, and this was probably due to unlabeled C02 in the leaf

chamber at the beginning of the labelling period, and to the release of low

specific activity C02 by the leaf respiration during the early hours of the

165

166

labelling period (Chapter 2). The specific activity of 14C02 could be monitored

with an ion chamber, or the relative amount of 14C02 in the gas stream could

be measured using a flow-through 14C detector. If two 14C02 sources were

used for C02 addition to the system under software control: one with a

relatively high 14C02 specific activity and one at the same 14C02 specific

activity that was to be maintained during the labelling period, a straightforward

modification of the software and hardware used for steady state labelling would

allow the specific activity of 14C02 to be maintained during the labelling period.

This would be accomplished by software comparison of the measured 14C02

specific activity or gas stream 14C content with the experimentally desired

14C02 specific activity or gas stream 14C content, followed by the addition of

high specific activity 14C02 whenever the specific activity 14C02 in the system

was lower than desired. The maintenance of a constant 14C02 specific activity

in the leaf chamber would probably decrease the amount of time required for

the specific activity of the leaf sugar pool to reach a steady state with the

specific activity of 14C02 used for labelling.

Empty seed coat methodology. The use of the empty seed coat

technique is limited by the ability to compare results from perfused seed coats

with results from intact seeds. The steady state labelling methods used in the

current study were adequate for quantitative comparisons of photosynthate

transport to perfused versus intact seeds, but the correction of perfused seed

coat data for the loss of transport capacity caused by removing a portion of the

seed coat was not precise (chapter 3). Seed size corrections based on seed

coat area appear to be more precise, although these methods are difficult and

time consuming (Gifford and Thorne, 1986; Hanson 1986a, 1988). In addition,

seed coat area measurements are based on the assumption that unloading

167

from the seed coat is uniform over the entire surface of the seed coat, which

may not be an accurate reflection of seed coat characteristics (Grusak and

Minchin, 1988; Chapter 3). The use of the empty seed coat technique could be

improved by the analysis of the distribution of seed coat unloading over the

surface of the seed coat.

Non-destructive methods for the analysis of sugar, 14C, and sugar

specific activity in seed coats during perfusion would be desirable for the

improvement of our understanding of photosynthate compartmentation within

the seed coat, and could enhance the resolution of photosynthate unloading

pathways within legume seed coats. Although tissue samples might be

obtained from seed coats during the perfusion period, it is unlikely that this

methodology would provide quantitative data, and might also disturb phloem

import or photosynthate unloading from the seed coat. Minchin and Grusak

(1988) and Grusak and Minchin (1988) used direct monitoring of 11C within

seed coats and perfusates to observe tracer compartmentation between

retention in the seed coat and release to a perfusion solution, but these

methods were far from quantitative. Destructive sampling of seed coats after

perfusion for different amounts of time may be the only method available for the

quantitative analysis of sugar and tracer compartmentation in seed coats

during perfusion. Although this method would require a great deal of effort

(many replications and many time samples), the potential for high resolution

kinetic analysis of photosynthate compartmentation within the seed coat may

be worth the effort.

The analysis of perfusates released from perfused seed coats was

limited by the methods used for the determination of the specific activity of

eluted sugar (Chapter 3). The use of ion exchange columns to remove non­

sugar compounds from perfusates before 14C determination could allow the

168

specific activity of sugar eluted from the seed coat to be compared directly with

the specific activity of sugar remaining in the seed coat after perfusion. The

comparison of eluted sugar specific activity with the specific activity of sugar

remaining in the seed coat after perfusion could improve the resolution of

sugar compartmentation in the seed coat. This might lead to the discovery of

sugar pools in the seed coat with different rates of exchange with the high

specific activity sugar imported by the phloem. An even more detailed analysis

of sugar elution and compartmentation within the seed coat might be obtained

by the separation and specific activity determination of different sugars in seed

coats and perfusates using HPLC. The analysis of amino acids, organic acids,

potassium, and other substances eluted from and remaining in perfused seed

coats would also improve our understanding of photosynthate unloading in

legume seed coats.

Directions for Future Research

Turgor-sensitive sugar uptake by developing cotyledons. Sucrose

uptake by developing soybean cotyledons is at least partially mediated by a

sucrose/proton symport coupled with an electrogenic H+ -ATPase (Lichtner

and Spanswick, 1981a, 1981b; Lin, 1985). There is indirect evidence that

sucrose uptake by developing legume cotyledons may be increased by low cell

turgor (Wolswinkel and Ammerlaan, 1986; Wolswinkel et aI., 1986), and thus

the active uptake of sucrose by legume cotyledons may be turgor-sensitive.

The electrophysiological methods developed by Lichtner and Spanswick

(1981a, 1981b) could be combined with osmoticum treatments, and thus the

turgor-sensitivity of sucrose/proton cotransport in developing legume

cotyledons could be determined. Active carrier-mediated sucrose uptake

appears to be saturated at in vivo sucrose concentrations, which may indicate

169

that the rate of sucrose uptake by the cotyledons may be limited by this

process (Lichtner and Spanswick, 1981b; Patrick, 1981). Therefore, the turgor­

sensitivity of sucrose uptake by legume cotyledons may be important to the

determination of the rate of seed fill in legume seeds.

Sucrose/proton cotransport mechanisms in seed coats. Based on

the effects of fusicoccin, K+ , orthovanadate, and ABA on the efflux of protons

and 14C-photosynthates from perfused seed coats, Van Bel and Patrick (1984)

have proposed that sucrose unloading from Phaseolus vulgaris seed coats is

mediated by an outward-directed sucrose/proton symport. The existence of

sucrose/proton cotransport mechanisms in the seed coat could be

convincingly demonstrated by conventional electrophysiological techniques,

and the role of sucrose/proton cotransport mechanisms in the regulation of

photosynthate unloading from legume seed coats could be evaluated. In

addition, the turgor-sensitivity of electrogenic proton pumping and

sucrose/proton cotransport could be demonstrated by the coupling of

electrophysiological methods with osmoticum treatments.

Evaluation of the effects of potassium, growth regulators and other

treatments. The quantitative methods developed in this study for the analysis

of photosynthate unloading in developing legume ~eeds may prove useful for

the investigation of the effects of potassium, growth regulators and other

treatments on photosynthate unloading from seed coats. The effects of ABA

and K+ on photosynthate unloading are significant, but have not been

evaluated on a quantitative basis. The use of EDTA in the current study

produced interesting results, and a greater understanding the effects of EDTA

on phloem import and unloading in perfused seed coats might be gained by

long-term perfusion of seed coats with EDTA and various concentrations of an

170

osmoticum, without changes in the make-up of the perfusion solution during

the perfusion period. Treatment with metabolic inhibitors and PCMBS may be

combined with quantitative measurements of photosynthate unloading from

seed coats in order to help clarify the energetics and carrier-mediation of

photosynthate unloading.

Determination of the plasmolytic effects of osmoticum treatments.

Treatment of seed coats with osmoticum concentrations greater than 200 mM

caused visible plasmolysis of cells at the inner surface of the seed coat (Thorne

and Rainbird, 1983; Hanson, 1986a), but this did not lead to significant

membrane damage (Patrick, 1984). The plasmolytic effects of osmoticum

treatments could be studied by microscopic methods (Robert Turgeon,

personal communication), and this could allow the determination of osmoticum

effects on symplastic continuity within the seed coat. In addition, the extent of

plasmolysis in various tissues within the seed coat following treatment with

different concentrations of an osmoticum could be used to evaluate the

concentrations of solutes within cells along the path of photosynthate

unloading from the phloem to the apoplast. The existence of a solute

concentration gradient within the cells of the seed coat from the phloem to the

inner surface of the seed coat would be expected for a symplastic route of

phloem unloading, and thus the microscopic analysis of seed coat cells

following plasmolytic treatments could be very useful for the study of

photosynthate unloading from seed coats.

Changes in photosynthate unloading during seed development.

The application of quantitative photosynthate unloading measurements to seed

coats at different stages of development would be useful, because there is

evidence that the characteristics of photosynthate unloading change during

171

seed development (Wolswinkel and Ammerlaan, 1986). A detailed analysis of

apoplastic solutes during seed development would also be very valuable from

a physiological perspective, as the concentration of solutes in the apoplast

may be a control point for the processes which regulate sink strength (Thorne,

1985; Wolswinkel, 1985b, 1988).

The use of the empty seed coat technique for crop improvement.

The empty seed coat technique has been used for the evaluation of genetic

associations between seed dry matter accumulation rates and seed coat

transport capacity (Hanson, 1986a, 1986b, 1988). Hanson (1986a, 1986b)

measured the rate of sucrose unloading from seed coats for one hour in the

presence of 2.5 mM EDTA, and found that these measurements of "sucrose

release rate" were not correlated with seed growth rates. EDTA may have

induced a general leakiness of sugars from the seed coat, and may thus have

clouded the analysis of seed coat transport capacity (Chapter 4). The

collection of sucrose eluted from seed coats during the first hour of perfusion is

not a reliable method for the analysis of phloem import and unloading in seed

coats, because attached and detached seed coats do not differ with respect to

sucrose elution during the first two hours of perfusion (Chapter 3). For both of

the reasons described above, Hanson's measurements were unlikely to be

useful for determination of genotypic differences in seed coat transport

capacity. The use of sucrose efflux measurements between two and eight

hours after the initiation of perfusion may provide a useful measure for

comparisons between genotypes with respect to seed coat transport capacity

(Chapter 3).

Concluding Remarks

The results of the current study demonstrate that the phloem import of

photosynthates to seeds, and the unloading of photosynthates from the seed

172

coat to the apoplast are regulated by turgor-sensitive mechanisms in

Phaseolus vulgaris (Chapters 3 and 4). These results confirm those of previous

studies, which indicate that a high concentration of solutes in the apoplast of

sink tissues is required to lower phloem turgor for the purpose of enhancing

phloem import by mass flow (Chapter 1, and reviews by: Wolswinkel, 1985b,

1988; Lang and Thorpe, 1986; Ho, 1988; Patrick, 1988). The regulation of

photosynthate unloading from seed coats by cell turgor may facilitate the

coordination of phloem import with photosynthate release from the seed coat

for uptake by the embryo. The stimulation of phloem import by low cell turgor is

balanced by a reduction in photosynthate unloading from the seed coat, and

these processes appear to function as a turgor homeostat for the maintenance

of a high concentration of photosynthates in the seed coat apoplast (Chapters

1 and 3; Patrick et aI., 1986). A high concentration of photosynthates in the

apoplast would increase the uptake and accumulation of photosynthates by

the developing embryo, which would thus favor a high rate of seed growth. The

water relations of seed tissues change during development, and changes in

osmotic and pressure potentials in seeds appear to be important for seed

development (Chapter 1). For all of the reasons discussed above, turgor­

sensitive mechanisms are important to seed growth and development.

There is a great deal of evidence in the literature for the regulation of

photosynthate production and partitioning by sink demand, and thus, the

potential exists for the improvement of seed yield in legumes by the

enhancement of the processes which regulate photosynthate partitioning in the

developing seed (Chapter 1). The mechanisms by which sink demand

influences photosynthate production and export to the sink are complex.

However, recent studies have demonstrated that the phloem loading of

173

sucrose in the leaf for export to sinks may be enhanced by low phloem turgor

(Chapter 1). The enhancement of phloem loading and export by low cell turgor

may provide a mechanism by which sink demand regulates export from leaves,

because the turgor differential from source to sink could be regulated by the

maintenance of a low phloem turgor in the sink (Chapter 1). Thus, the

demonstration of turgor-sensitive photosynthate unloading mechanisms in

developing legume seeds may provide a basis for the improvement of

photosynthate partitioning to seeds. Although the potential exists for the

enhancement of seed yield by the manipulation of photosynthate unloading

mechanisms by genetic or other means, it must be understood that the

mechanisms which regulate whole plant photosynthate partitioning are plastic,

and the manipulation of one process in the plant may be compensated for by

changes in the capacity of other processes (Chapter 1).

The agronomic potential of the techniques developed in this study.

The use of steady state labelling methods in the field is not possible at the

current time, and the relative expense and labor involved in this technique will

not allow comparisons between many genotypes without a tremendous

commitment to this task. However, sucrose efflux measurements on a large

scale in the greenhouse or in the field are possible, and this may be a useful

goal for the evaluation of the empty seed coat technique as a method for the

study of genotypic differences in seed coat transport capacity. The potential for

a genetic association between seed coat transport capacity and the rate of

seed fill, or total seed yield is probably limited by the many different partitioning

mechanisms that contribute to the total seed yield of grain legumes. The

processes that determine the total seed number or the duration of seed fill in a

particular genotype may be more important to seed yield than the rate of seed

174

fill, and the rate of seed fill may be determined by factors other than the rate of

photosynthate unloading from the seed coat (such as the number of cells in

the cotyledons, or the efficiency of photosynthate uptake and storage by the

cotyledons). Therefore, the agronomic importance of quantitative methods for

the analysis of photosynthate unloading in developing legume seeds may lie in

the investigation of the physiology of photosynthate partitioning mechanisms,

which may lead to an understanding of these mechanisms on a molecular

level. In particular, an understanding of the molecular basis for turgor-sensitive

efflux control mechanisms in seed coats (and perhaps in the embryo) of

developing legume seeds could facilitate the manipulation of these

mechanisms by the methods of molecular biology.

LITERATURE CITED

Ackerson RC (1984) Regulation of soybean embryogenesis by abscisic acid. J

Exp Bot 35:404-413

Aloni B, Wyse RE, Griffith S (1986) Sucrose transport and phloem unloading

in stem of Vicia faba: possible involvement of a sucrose carrier and

osmotic regulation. Plant Physiol 81 :482-486

Aloni B, Daie, J, Wyse RE (1988) Regulation of apoplastic pH in source leaves

of Vicia faba by gibberellic acid. Plant Physiol 88:367-369

Ashley, DA, Boerma, HR (1989) Canopy photosynthesis and its association

with seed yield in advanced generations of a soybean cross. Crop Sci

29: 1 042-1 045

Azcon-Bieto J (1983) Inhibition of photosynthesis by carbohydrates in wheat

leaves. Plant Physiol 73:681-686

Bennett AB, Spanswick RM (1983) Derepression of amino acid-H + cotransport in developing soybean embryos. Plant PhysioI72:781-786

Bergmeyer HU, Bernt E (1974) Determination of D-glucose with glucose

oxidase and peroxidase. In HU Bergmeyer, ed, Methods of Enzymatic

Analysis, Vol 3. Verlag Chemie, Weinheim/ Academic Press, New York,

pp 1205-1215

Briggs CL, Westoby M, Selkirk PM, Oldfield RJ (1987) Embryology of early

abortion due to limited maternal resources in Pisum sativum L. Ann Bot

59:611-619

Briskin DP (1986) Sucrose transport in isolated tonoplast vesicles from sugar

beet taproot. In J Cronshaw, WJ Lucas, RJ Giaquinta, eds, Phloem

Transport, Plant Biology, Vol 1. AR Liss, New York, pp 111-123

175

176

Brun WA, Brenner ML, Schussler J (1986) Translocation of abscisic acid and

assimilates in soybean. In J Cronshaw, WJ Lucas, RJ Giaquinta, eds,

Phloem Transport, Plant Biology, Vol 1. AR Liss, New York, pp 589-596

Carlson DR, Brun WA (1984) Effect of shotened photosynthetic period on

14C-assimilate translocation and partitioning in reproductive soybeans.

Plant Physiol 75:881-886

Carlson DR, Brun WA (1985) Effect of the reproductive sink on 14C-assimilate

partitioning between starch and water-soluble compounds in soybean

leaves. Crop Sci 25:283-288

Carr DJ, Skene KGM (1961) Diauxic growth curves in seeds, with special

reference to French beans (Phaseolus vulgaris). Aust J Bioi Sci 14:1-12

Clifford PE, Offler CE, Patrick JW (1986) Growth regulators have rapid

effects on photosynthate unloading from seed coats of Phaseolus

vulgaris L. Plant Physiol 80:635-637

Clifford PE, Offler CE, Patrick JW (1987) Injection of growth regulators into

seeds growing in situ on plants of Phaseolus vulgaris with a double-fruit

sink system. Can J Bot 65:612-615

Clough JM, Peet MM, Kramer PJ (1981) Effects of high atmospheric C02

and sink size on rates of photosynthesis of a soybean cultivar. Plant

PhysioI67:1007-1010

Cook H, Oparka KJ (1983) Movement of fluorescein into isolated caryopses of

wheat and barley. Plant, Cell Environ 6:239-242

Cornish K, Spanswick RM (1987) Amino acid:H+ cotransport by developing

cotyledons of pea and soybean (abstract No. 629). Plant PhysioI83:S-

104

177

Daie J (1987a) Interaction of cell turgor and hormones on sucrose uptake in isolated phloem of celery. Plant Physiol 84: 1 033-1 037

Daie J (1987b) Sucrose uptake in isolated phloem of celery is a single

saturable transport system. Planta 171:474-482

De Ruiter H, Schuurmans J, Kolloffel C (1984) Amino acid leakage from

cotyledons of developing and germinating pea seeds. J Plant Physiol

116:47-57

Delrot S (1987) Phloem loading: apoplastic or symplastic? Plant Physiol

Biochem 25:667-676

Delrot S, Bonnemain J-L (1985) Mechanism and control of Phloem Transport,

Plant Biology, Vol 1. Physiol Veg 23:199-220

Dunphy EJ, Hanway JJ, Green DE (1979) Soybean yields in relation to days between specific developmental stages. Agron J 71:917-920

Egli DB (1988) Alterations in plant growth and dry matter distribution in

soybean. Agron J 80:86-90

Egli DB, Fraser J, Leggett JE, Poneleit CG (1981) Control of seed growth in

soya beans (Glycine max (L.) Merr.). Ann Bot 48:171-176

Egli DB, Guffy RD, Meckel LW, Leggett JE (1985) The effect of source-sink

alterations on soybean seed growth. Ann Bot 55:395-402

Egli DB, Leggett JE (1976) Rate of dry matter accumulation in soybean seeds

with varying source-sink ratios. Agron J 68:371-373

Egli DB, Leggett JE, Wood JM (1978) Influence of soybean seed size and

position on the rate and duration of filling. Agron J 70: 127-130

178

Egli DB, orr JH, Pfeifer TW (1984) Genotypic variation for duration of seedfill

in soybean. Crop Sci 24:587-592

Ellis EC, Spanswick RM (1987) Sugar efflux from attached seed coats of

Glycine max (L.) Merr. J Exp Bot 38:1470-1483

Estruch, JJ, Pereto, JG, Vercher, Y, Beltran, JP (1989) Sucrose loading in

isolated veins of Pisum sativum: regulation by abscisic acid, gibberellic

acid, and cell turgor. Plant Physiol 91 :259-265

Fader GM, Koller HR (1985) Seed growth rate and carbohydrate pool sizes of

the soybean fruit. Plant Physiol 79:663-666

Felker FC, Shannon JC (1980) Movement of 14C-labeled assimilates into

kernels of Zea mays L. III. An anatomical examination and

microautoradiographic study of assimilate transfer. Plant Physiol 65:864-

870

Fisher DB (1970a) Kinetics of C-14 translocation in soybean. I. Kinetics in the

stem. Plant Physiol 45: 1 07-113

Fisher DB (1970b) Kinetics of C-14 translocation in soybean. II. Kinetics in the

leaf. Plant PhysioI45:114-118

Fisher DB (1970c) Kinetics of C-14 translocation in soybean. III. Theoretical

considerations. Plant Physiol 45: 119-125

Fisher DB (1978) An evaluation of the Munch hypotheSiS for Phloem

Transport, Plant Biology, Vol 1 in soybean. Planta 139:25-28

Fisher DB, Gifford RM (1986) Accumulation and conversion of sugars by

developing wheat grains. VI. Gradients along the transport pathway

from the peduncle to the endosperm cavity during grain filling. Plant

Physiol 82: 1 024-1 030

179

Flinn AM, Pate JS (1970) A quantitative study of carbon transfer from pod and

subtending leaf to the ripening seeds of the field pea (Pisum arvense L.)

J Exp Bot 21 :71-82

Fondy BR, Geiger DR (1980) Effect of rapid changes in sink-source ratio on

export and distribution of products in leaves of Beta vulgaris L. and

Phaseolus vulgaris L.. Plant Physiol 66:945-949

Fondy BR, Geiger DR (1982) Diurnal pattern of translocation and

carbohydrate metabolism in source leaves of Beta vulgaris L. Plant

Physiol 70:671-676

Fondy BR, Geiger DR (1983) Contrel of export and partitioning among sinks

by allocation of products of photosynthesis in source leaves. In DD

Randall, D Blevins, RL Larson, B Rapp, eds, Current Topics in Plant

Biochemistry and Physiology, Vol 2. University of Missouri at Columbia,

Columbia, pp 33-42

Fondy BR, Geiger DR, Servaites JC (1989) Photosynthesis, carbohydrate

metabolism, and export in Beta vulgaris L. and Phaseolus vulgaris L. during square and sinusoidal light regimes. Plant Physiol 89:396-402

Foyer CH (1988) Feedback inhibition of photosynthesis through source-sink

regulation in leaves. Plant Physiol Biochem 26:483-492

Franceschi VR (1986) Temporary storage and its role in partitioning among

sinks. In J Cronshaw, WJ Lucas, RJ Giaquinta, eds, Phloem Transport,

Plant Biology, Vol 1. AR Liss, New York, pp 399-409

Geiger DR (1980) Measurement of translocation. Methods Enzym 69:561-571

Geiger DR (1986) Processes affecting carbon allocation and partitioning

among sinks. In J Crenshaw, WJ Lucas, RJ Giaquinta, eds, Phloem

Transport, Plant Biology, Vol 1. AR Liss, New York, pp 375-388

180

Geiger DR (1987) Understanding interactions of source and sink regions in plants. Plant Physiol Biochem 25:659-666

Geiger DR, Fondy BR (1980) Response of phloem loading and export to rapid

changes in sink demand. Ber Deutsch Bot Ges Bd 93: 177-186

Geiger DR, Ploeger BJ, Fox TC, Fondy BR (1983) Sources of sucrose translocated from illuminated sugar beet source leaves. Plant Physiol

72:964-970

Geiger DR, Shieh WJ (1988) Analysing partitioning of recently fixed and of

reserve carbon in reproductive Phaseolus vulgaris L. plants. Plant, Cell Environ 11 :777-783

Geiger DR, Shieh WJ, Saluke RM (1989) Carbon partitioning among leaves, fruits, and seeds during development of Phaseolus vulgaris L.. Plant

Physiol 91 :291-297

Gent MPN, Kiyomoto RK (1989) Assimilation and distribution of photosynthate in winter wheat cultivars differing in harvest index. Crop

Sci 29:120-125

Gifford RM (1986) Partitioning of photoassimilate in the development of crop yield. In J Cronshaw, WJ Lucas, RJ Giaquinta, eds, Phloem Transport, Plant Biology, Vol 1. AR Liss, New York, pp 535-549

Gifford RM, Evans LT (1981) Photosynthesis, carbon partitioning and yield. Ann Rev Plant Physiol 32:485-509

Gifford RM, Thorne JH (1985) Sucrose concentration at the apoplastic interlace between seed coat and cotyledons of developing soybean seeds. Plant Physiol 77:863-868

Gifford RM, Thorne JH (1986) Phloem unloading in soybean seed coats: dynamics and stability of efflux into attached 'empty ovules'. Plant Physiol 80:464-469

181

Gifford RM, Thorne JH, Hitz WD, Giaqulnta RT (1984) Crop productivity and I

photoassimilate partitioning. Science 225:801-808

Goeschl JD, Magnuson CE, DeMichele DW, Sharpe PJH (1976)

Concentration-dependent unloading as a necessary assumption for a

closed form mathematical model of osmotically driven pressure flow in

the phloem. Plant Physiol 58:556-562

Gordon AJ (1986) Diurnal patterns of photosynthate allocation and partitioning

among sinks. In J Cronshaw, WJ Lucas, RJ Giaquinta, eds, Phloem

Transport, Plant Biology, Vol 1. AR Liss, New York, pp 499-517

Grange RI (1985) Carbon partitioning and export in mature leaves of pepper

(Capsicum annuum). J Exp Bot 36:734-744

Griffith SM, Jones RJ, Brenner ML (1986) characteristics of in vitro sugar

transport by developing maize kernels. In J Cronshaw, WJ Lucas, RJ

Giaquinta, eds, Phloem Transport, Plant Biology, Vol 1. AR Liss, New

York, pp 283-284

Griffith SM, Jones RJ, Brenner ML (1987a) In vitro sugar transport in Zea

mays L. kernels. I. Characteristics of sugar absorption and metabolism

by developing maize endosperm. Plant Physiol 84:467-471

Griffith SM, Jones RJ, Brenner ML (1987b) In vitro sugar transport in Zea

mays L. kernels. II. Characteristics of sugar absorption and metabolism

by isolated developing embryos. Plant PhysioI84:472-475

Groussol J, Delrot S, Caruhel P, Bonnemain J-L (1986) Design of an

improved exudation method for phloem sap collection and its use for the

study of phloem mobility of pesticides. Physiol Veg 24: 123-133

Grusak MA, Minchin PEH (1988) Seed coat unloading in Pisum sativum -

osmotic effects in attached versus excised empty ovules. J Exp Bot

39:543-559

182

Guldan SJ, Brun WA (1987) Relationship of tissue water relations to

asparagine uptake in developing soybean seeds. Crop Sci 27:720-725

Hanson WD (1985) Association of seed yield with partitioned lengths of the

reproductive period in soybean genotypes. Crop Sci 25:525-529

Hanson WD (1986a) Rates of sucrose release from opened seed coats of

soybean as affected by genotypes: relation with rate of seed

development. Crop Sci 26:301-307

Hanson WD (1986b) Control of dry matter accumulation in soybean seeds.

Crop Sci 26: 1195-1200

Hanson WD (1988) Analysis of differences in sink activity among soybean

genotypes based on dry matter accumulation rates per unit seedcoat

area. Crop Sci 28:830-834

Hanson WD, Kenny ST (1985) Genotypic differences in soybean affecting the

rates of assimilate transport from the leaf. Crop Sci 25:229-234

Hardham AR (1976) Structural aspects of the pathways of nutrient flow to the

developing embryo and cotyledons of Pisum sativum L. Aust J Bot

24:711-721

Hayes PM, Patrick JW, Offler CE (1987) The cellular pathway of radial

transfer of photosynthates in stems of Phaseolus vulgaris L.: effects of

cellular plasmolysis and p-chloromercuribenzene sui phonic acid. Ann

Bot 59:635-642

Heitholt JJ, Egli DB, Leggett JE, MacKown CT (1986) Role of assimilate and

carbon-14 photosynthate partitioning in soybean reproductive abortion.

Crop Sci 26:999-1004

183

Herold A (1980) Regulation of photosynthesis by sink activity - the missing link. New Phytol 86:131-144

Herzog H (1982) Relation of source and sink during grain filling period in wheat

and some aspects of its regulation. Physiolog Plant 56: 155-160

Ho LC (1986) Metabolism and compartmentation of translocates in sink organs. In J Cronshaw, WJ Lucas, RJ Giaquinta, eds, Phloem

Transport, Plant Biology, Vol 1. AR Uss, New York, pp 317-324

Ho LC (1988) Metabolism and compartmentation of imported sugars in sink organs in relation to sink strength. Ann Rev Plant Physiol Plant Mol Bioi

39:355-378

Ho LC, Gifford RM (1984) Accumulation and conversion of sugars by developing wheat grains V. The endosperm apoplast and apoplastic transport. J Exp Bot 35:58-73

Hoddinott J, Jolliffe P (1988) The influence of elevated carbon dioxide concentrations on the partitioning of carbon in source leaves of Phaseolus vulgaris. Can J Bot 66:2396-2401

Hsu FC (1979) A developmental analysis of seed size in common bean. Crop

Sci 19:226-230

Hsu Fe, Bennett AB, Spanswick RM (1984) Concentrations of sucrose and nitrogenous compounds in the apoplast of developing soybean seed coats and embryos. Plant PhysioI75:181-186

Huber SC, Kerr PS (1987) Regulation of photosynthetic starch/sucrose metabolism. In OW Newman, KG Wilson, eds, Models in Plant Physiology and Biochemistry, Vol 1. CRC Press, Boca Raton, pp 79-83

184

Huber SC, Kerr PS, Kalt-Torres W (1986) Biochemical control of allocation of

carbon for export and storage in source leaves. In J Cronshaw, WJ

Lucas, RJ Giaquinta, eds, Phloem Transport, Plant Biology, Vol 1. AR

Liss, New York, pp 355-367

Hughes JS, Swanson BG (1985) Microstructural changes in maturing seeds

of the common bean (Phaseolus vulgaris L.). Food Microstruct 4:183-

189

Jaeger CH, Goeschl JD, Magnuson CE, Fares Y, Strain BR (1988) Short­

term responses of phloem transport to mechanical perturbation. Physiol

Plant 72:588-594

Jenner CF, Rathjen AJ (1978) Physiological basis of genetic differences in the

growth of grains of six varieties of wheat. Aust J Plant Physiolog 5:249-

262

Kaiser WM, Martinoia E (1985) Absence of an apoplastic step in assimilate

transport to the phloem? A comparison of assimilate efflux from leaf

slices, mesophyll protoplasts and a unicellular green alga. J Plant

Physiol121 :463-474

Kallarackal J, Milburn JA (1984) Specific mass transfer and sink-controlled

phloem translocation in castor bean. Aust J Plant Physiol11 :483-490

Keener ME, DeMichele OW, Sharpe PJH (1979) Sink metabolism: a

conceptual framework for analysis. Ann Bot 44:659-669

Kinraide TB, Wyse RE (1986) Electrical evidence for turgor inhibition of proton

extrusion in sugar beet taproot. Plant Physiol 86: 1148-1150

Kouchi H, Yoneyama T (1984a) Dynamics of carbon photosynthetically

assimilated in nodulated soya bean plants under steady-state conditions

1. Development and application of 13C02 assimilation system at a

constant 13C abundance. Ann Bot 53:875-882

185

Kouchi H, Yoneyama T (1984b) Dynamics of carbon photosynthetically

assimilated in nodulated soya bean plants under steady-state conditions

2. the incorporation of 13C into carbohydrates, organic acids, amino

acids, and some storage compounds. Ann Bot 53:883-896

Kursanov AL (1984) Endogenous regulation of assimilate transport and

source-sink relations in plants. Fiziologiya Rastenii (Engl Transl) 31 :579-

595

Lanfermeijer F (1987) Changes in valine-uptake during pea-seed development

(abstract No. 923). Plant PhysioI83:S-923

Lanfermeijer FC, Koerselman-Kooji JW, Kolloffel C, Borstlap AC (1989)

Release of amino acids from cotyledons of developing seeds of pea

(Pisum sativum L.). J Plant PhysioI134:592-597

Lang A (1983) Turgor-regulated translocation. Plant, Cell Environ 6:683-689

Lang A, Thorpe MR (1983) Analysing partitioning in plants. Plant, Cell Environ

6:267-274

Lang A, Thorpe MR (1986) Water potential, translocation and assimilate

partitioning. J Exp Bot 37:495-503

Lang A, Thorpe MR, Edwards WRN (1986) Plant water potential and

translocation. In J Cronshaw, WJ Lucas, RJ Giaquinta, eds, Phloem

Transport, Plant Biology, Vol 1. AR Liss, New York, pp 193-194

Le Page-Degivry MT, Barthe P, Prevost I, Boulon B (1989) Regulation of

abscisic acid translocation during embryo maturation of Phaseolus

vulgaris. Physiol Plant 77:81-86

Lenton JR (1984) Are plant growth substances involved in the partitioning of

assimilate to developing reproductive sinks? Plant Growth Reg 2:267-

276

186

Lichtner FT, Spanswick RM (1981a) Electrogenic sucrose transport in

developing soybean cotyledons. Plant PhysioI67:869-874

Lichtner FT, Spanswick RM (1981 b) Sucrose uptake by developing soybean

cotyledons. Plant Physiol 68:693-698

Lin W (1985) Linear sucrose transport in protoplasts from developing soybean

cotyledons. Plant PhysioI78:649-651

Lingle SE, Chevalier P (1984) Movement and metabolism of sucrose in

developing barley kernels. Crop Sci 24:315-319

Lingle SE, Chevalier P (1985) Development of the vascular tissue of the wheat

and barley carypsis as related -to the rate and duration of grain filling.

Crop Sci 25: 123-128

Lopez R, Dathe W, Sembdner G (1989) Abscisic acid in different parts of the

developing soybean fruit. Biochem Physiol Pflanzen 184:267-276

Lucas EO, Milbourn GM, Whitford PN (1976) The translocation of 14C

photosynthate from leaves and pods in Phaseolus vulgaris. Ann Appl

Bioi 83:285-290

Maness NO, McBee GG (1986) Role of placental sac in endosperm

carbohydrate import in sorghum caryopses. Crop Sci 26: 1201-1207

Marowitch J, Richter C, Hodinott J (1986) The influence of plant temperature

on photosynthesis and translocation rates in bean and soybean. Can J

Bot 64:2337-2342

Martinez-Carrasco R, Perez P, Martin del Molino I, Ulloa M, Rojo B (1988)

Regulation of grain weight by supply of assimilates and starch granule

development in three winter wheat varieties. J Exp Bot 39:1723-1733

187

Meckel L, Egli DB, Phillips RE, Radcliffe D, Leggett JE (1984) Effect of

moisture stress on seed growth in soybeans. Agron J 76:647-650

Minchin PEH, Grusak MA (1988) Continuous in vivo measurement of carbon

partitioning within whole plants. J Exp Bot 39:561-571

Minchin PEH, McNaughton GS (1986) Phloem unloading within the seed coat

of Pisum sativum observed using surgically modified seeds. J Exp Bot

37:1151-1163

Minchin PEH, Ryan KG, Thorpe MR (1984) Further evidence of apoplastic

unloading into the stem of bean: identification of the phloem buffering

pool. J Exp Bot 35: 1744-1753

Minchin PEH, Thorpe MR (1987a) Is phloem transport due to a hydrostatic

pressure gradient? Supporting evidence from pressure chamber

experiments. Aust J Plant Physiol 14:397-402

Minchin PEH, Thorpe MR (1987b) Measurement of unloading and reloading

of photo-assimilate within the stem of bean. J Exp Bot 38:211-220

Minchin PEH, Thorpe MR (1989) Carbon partitioning to whole versus

surgically modified ovules of pea: an application of the in vivo

measurement of carbon flows over many hours using the short-lived

isotope carbon-11. J Exp Bot 40:781-787

Mullen JA, Koller HR (1988a) Daytime and nighttime carbon balance and

assimilate export in soybean leaves at different photon flux densities.

Plant Physiol 86:880-884

Mullen JA, Koller HR (1988b) Trends in carbohydrate depletion, respiratory

carbon loss, and assimilate export from soybean leaves at night. Plant

PhysioI86:517-521

Munch, E (1930) Die stoffbewegungen in der pflanze. Gustaf Fischer, Jena

188

Murphy R (1986) Symplastic sieve tube unloading in seed coats of Phaseolus

vulgaris is a necessary requirement for pressure-driven translocation. In

J Cronshaw, WJ Lucas, RJ Giaquinta, eds, Phloem Transport, Plant

Biology, Vol 1. AR Liss, New York, pp 259-262

Murray DR (1987) Nutritive role of seedcoats in developing legume seeds.

Amer J Bot 74:1122-1137

Nelson CJ (1988) Genetic associations between photosynthetic

characteristics and yield: review of the evidence. Plant Physiol Biochem

26:543-554

Ntsika G, Delrot S (1986) Changes in apoplastic and intracellular leaf sugars

induced by the blocking of export in Vicia faba. Physiol Plant 68:145-153

Offler CE, Nerlich SM, Patrick JW (1989) Pathway of photosynthate transfer

in the developing seed of Vicia faba L. transfer in relation to seed

anatomy. J Exp Bot 40:769-780

Offler CE, Patrick JW (1984) Cellular structures, plasma membrane surface

areas and plasmodesmatal frequencies of seed coats of Phaseolus vulgaris L. in relation to photosynthate transfer. Aust J Plant Physiol

11:79-99

Offler CE, Patrick JW (1986) Cellular pathway and hormonal control of short­

distance transfer in sink regions. In J Cronshaw, WJ Lucas, RJ Giaquinta, eds, Phloem Transport, Plant Biology, Vol 1. AR Liss, New

York, pp 295-306

Oliker M, Poljakoff-Mayber A, Mayber AM (1978) Changes in weight, nitrogen accumulation, respiration, and photosynthesis during growth

and development of seeds and pods of Phaseolus vulgaris. Amer J Bot 65:366-371

189

Oparka KJ (1986) Phloem unloading in the potato tuber. Pathways and sites of

ATPase. Protoplasma 131:201-210

Oparka KJ, Gates P (1984) Sink anatomy in relation to solute movement in

rice (Qryza sativa L.): a summary of findings. Plant Growth Reg 2:297-

307

Oparka KJ, Prior DAM (1987) 14C sucrose efflux from the perimedulla of

growing potato tubers. Plant, Cell Environ 10:667-675

Oparka KJ, Prior DAM (1988) Movement of lucifer yellow CH in potato tuber

storage tissues: a comparison of symplastic and apoplastic transport.

Planta 176:533-540

Oparka KJ, Wright KM (1988a) Influence of cell turgor on sucrose partitioning

in potato tuber storage tissues. Planta 175:520-526

Oparka KJ, Wright KM (1988b) Osmotic regulation of starch synthesis in

potato tubers? Planta 174: 123-126

Patrick JW (1981) An in vitro assay of sucrose uptake by developing bean

cotyledons. Aust J Plant Physiol 8:221-235

Patrick JW (1983) Photosynthate unloading from seed coats of Phaseolus

vulgaris L.. General characteristics and facilitated transfer. Z

Pflanzenphysiol 111 :9-18

Patrick JW (1984) Photosynthate unloading from seed coats of Phaseolus

vulgaris L.. Control by tissue water relations. J Plant PhysioI115:297-

310

Patrick JW (1987) Effect of potassium on photosynthate unloading from seed

coats of Phaseolus vulgaris L.. Specificity and membrane location. Ann

Bot 59:181-190

190

Patrick JW (1988) Assimilate partitioning in relation to crop productivity.

Hortscience 23:33-40

Patrick JW, McDonald R (1980) Pathway of carbon transport within

developing ovules of Phaseolus vulgaris L. Aust J Plant Physiol 7:671-

684

Patrick JW, Mulligan DR (1989) Gibberellic-acid-promoted transport of

assimilates in stems of Phaseolus vulgaris. Effects on assimilate

accumulation by the sink. Ann Bot 63:581-587

Patrick JW, Jacobs E, Offler CE, Cram WJ (1986) Photosynthate unloading

from seed coats of Phaseolus vulgaris L.- nature and cellular location of

turgor-sensitive unloading. J Exp Bot 37:1006-1019

Peet MM, Bravo A, Wallace DH, Ozbun JL (1977) Photosynthesis, stomatal

resistance, and enzyme activities in relation to yield of field-grown dry

bean varieties. Crop Sci 17:287-293

peoples MB, Pate JS, Atkins CA, Murray, DR (1985) Economy of water,

carbon, and nitrogen in the developing cowpea fruit. Plant Physiol

77:142-147

Pereto JG, Beltran, JP (1987) Hormone-directed sucrose transport during

fruit set induced by gibberellins in Pisum sativum. Physiol Plant 69:356-

360

Perry CA, Leigh RA, Tomos AD, Wyse RE, Hall JL (1987) The regulation of

turgor pressure during sucrose mobilization and salt accumulation by

excised storage-root tissue of red beet. Planta 170:353-361

Pfeiffer TW, Egli DB (1988) Heritability of seed-filling period estimates in

soybean. Crop Sci 28:921-925

191

Plaut Z, Mayoral ML, Reinhold L (1987) Effect of altered sink:source ratio on

photosynthetic metabolism of source leaves. Plant PhysioI85:786-791

Porter GA, Knievel DP, Shannon JC (1985) Sugar efflux from maize (Zea

mays L.) pedicel tissue. Plant PhysioI77:524-531

Porter GA, Knievel DP, Shannon JC (1987a) Assimilate unloading from maize

(Zea mays L.) pedicel tissues I. Evidence for regulation of unloading by

cell turgor. Plant PhysioI83:131-136

Porter GA, Knievel DP, Shannon JC (1987b) Assimilate unloading from maize

(Zea mays L.) pedicel tissues II. Effects of chemical agents on sugar,

amino acid, and 14C-assimilate unloading. Plant PhysioI85:558-565

Rasmusson DC (1987) An evaluation of ideotype breeding. Crop Sci 27: 1140-

1146

Rocher, JP (1988) Comparison of carbohydrate compartmentation in relation

to photosynthesis, assimilate export and growth in a range of maize

genotypes. Aust J Plant Physiol 15:677-686

Ross GS, Minchin PEH, McWha JA (1987) Direct evidence of abscissic acid

affecting phloem unloading within the seed coats of peas. J Plant

PhysioI129:435-441

Saab IN, Obendorf RL (1989) Soybean seed water relations during in situ and

in vitro growth and maturation. Plant PhysioI89:610-616

Sakri FAK, Shannon JC (1975) Movement of 14C-labeled sugars into kernels

of wheat ITriticum aestivum L.). Plant PhysioI55:881-889

Schmalstig JG, Geiger DR (1985) Phloem unloading in developing leaves of

sugar beet. I. Evidence for pathway through the symplast. Plant Physiol

79:237-241

192 ,

Schmalstig JG, Geiger DR, Hitz WD (1986) Pathway of phloem unloading in

developing leaves of sugar beet. In J Cronshaw, WJ Lucas, RJ

Giaquinta, eds, Phloem Transport, Plant Biology, Vol 1. AR Liss, New

York, pp 231-237

Schussler JR, Brenner ML, Brun WA (1984) Abscissic acid and its

relationship to seed filling in soybeans. Plant Physiol 76:301-306

Servaites JC, Geiger DR, Tucci MA, Fondy BR (1989) Leaf carbon

metabolism and metabolite levels during a period of sinusoidal light.

Plant Physiol 89:403-408

Shannon JC, Porter GA, Knievel DP (1986) Phloem unloading and transfer of

sugars into developing corn endosperm. In J Cronshaw, WJ Lucas, RJ

Giaquinta, eds, Phloem Transport, Plant Biology, Vol 1. AR Liss, New

York, pp 265-277

Shaw AF, Grange RI, Ho LC (1986) The regulation of source leaf assimilate

compartmentation. In J Cronshaw, WJ Lucas, RJ Giaquinta, eds,

Phloem Transport, Plant Biology, Vol 1. AR Liss, New York, pp 391-398

Smith JG (1973) Embryo development in Phaseolus vulgaris II. Analysis of

selected inorganic ions, ammonia, organic acids, and sugars in the

endosperm liquid. Plant Physiol 51 :454-458

Smith JR, Nelson RL, Vasilas BL (1988) Variation among soybean breeding

lines in relation to yield and seed-fill duration. Agron J 80:825-829

Snedecor GW, Cochran WG (1980) Statistical Methods, Ed 7. Iowa State

University Press, Ames

Snyder FW, Carlson GE (1984) Selecting for partitioning of photosynthetic

products in crops. Adv Agron 37:47-72

193

Spaeth SC, Randall HC, Sinclair TR, Vendeland JS (1984) Stability of

soybean harvest index. Crop Sci 76:482-486

Spaeth SC, Sinclair TR (1984) Soybean seed growth II. Individual seed mass

and component compensation. Agron J 76: 128-133

Stitt M (1986) Regulation of photosynthetic sucrose synthesis: integration,

adaptation, and limits. In J Cronshaw, WJ Lucas, RJ Giaquinta, eds,

Phloem Transport, Plant Biology, Vol 1. AR Uss, New York, pp 331-347

Sun 0, Wimmers LE, Turgeon R (1988) Scintillation counting of 14C-labelled

soluble and insoluble compounds in plant tissue. Anal Biochem

169:424-427

Sung SS, Xu O-P, Black, CC (1989) Identification of actively filling sucrose

sinks. Plant Physiol 89: 1117-1121

Thomas JF, Raper CO (1976) Photoperiodic control of seed filling for

soybeans. Crop Sci 16:667-672

Thorne JH (1979) Assimilate redistribution from soybean pod walls during

seed development. Agron J 71 :812-816

Thorne JH (1980) Kinetics of 14C-photosynthate uptake by developing

soybean fruit. Plant Physiol 65:975-979

Thorne JH (1981) Morphology and ultrastructure of maternal seed tissues of

soybean in relation to the import of photosynthates. Plant Physiol

67:1016-1025

Thorne JH (1982a) Temperature and oxygen effects on 14C-photosynthate

unloading and accumulation in developing soybean seeds. Plant Physiol

69:48-53

194

Thorne JH (1982b) Characterization of the active sucrose transport system of

immature soybean embryos. Plant Physiol 70:953-958

Thorne JH (1985) Phloem unloading of C and N assimilates in developing

seeds. Ann Rev Plant PhysioI36:317-343

Thorne JH (1986) Sieve tube unloading. In J Cronshaw, WJ Lucas, RJ

Giaquinta, eds, Phloem Transport, Plant Biology, Vol 1. AR Liss, New

York, pp 211-224

Thorne JH, Giaquinta RT (1984) Pathways and mechanisms associated with

carbohydrate translocation in plants. In DH Lewis, ed, Storage

Carbohydrates in Vascular Plants: Distribution, Physiology, and

Metabolism, Seminar Series, Society for Experimental Biology Vol 19.

Cambridge University Press, New York, pp 75-96

Thorne, JH, Koller, HR (1974) Influence of assimilate demand on

photosynthesis, diffusive resistances, translocation, and carbohydrate

levels of soybean leaves. Plant Physiol 54:201-207

Thorne JH, Rainbird RM (1983) An in vivo technique for the study of phloem

unloading in seed coats of developing soybean seeds. Plant Physiol

72:268-271

Thorpe MR, Minchin PEH (1988) Phloem loading and transport of

endogenously or exogenously labelled photo-assimilate in bean, beet,

maize, and cucurbit. J Exp Bot 39:1709-1721

Van Bel AJE, Patrick JW (1984) No direct linkage between proton-pumping

and photosynthate unloading from seed coats of Phaseolus vulgaris L.

Plant Growth Regulation 2:319-326

Vassey TL, Sharkey, TO (1989) Mild water stress of Phaseolus vulgaris plants

leads to reduced starch synthesis and extractable sucrose phosphate

synthase activity. Plant Physiol 89: 1 066-1 070

195

VerNooy CD, Thorne JH, Un W, Rainbird RM (1986) Cessation of assimilate

uptake in maturing soybean seeds. Plant Physiol 82:222-225

Walbot V, Clutter M, Sussex 1M (1972) Reproductive development and

embryogeny in Phaseolus. Phytomorphology 22:59-68

Wang TL, Smith CM, Cook SK, Ambrose MJ, Hedley CL (1987) An analysis

of seed development in Pisum sativum III. The relationship between the r locus, the water content and the osmotic potential of seed tissues in vivo

and in vitro. Ann Bot 59:73-80

Wann M, Raper CD (1987) Long-distance transport in phloem - a

mathematical simulation model. In OW Newman, KG Wilson, eds,

Models in Plant Physiology and Biochemistry, Vol 3. CRC Press, Boca

Raton, pp 25-28

Warren-Wilson J (1972) Control of crop processes. In AR Rees, KE Cockshull,

OW Hand, RG Hurd, eds, Crop Processes in Controlled E:nvironments.

Academic Press, London, pp 7-30

Wells R, Schulze LL, Ashley DA, Boerma HR, Brown RH (1982) Cultivar

differences in canopy apparent photosynthesis and their relationship to

seed yield in soybeans. Crop Sci 22:886-890

Wien HC, Altschuler SL, Ozbun JL, Wallace DH (1976) 14C-assimilate

distribution in Phaseolus vulgaris L. during the reproductive period. J

Amer Soc Hort Sci 101:510-513

Wolswinkel P (1985a) Effect of inhibitors on solute efflux from seed-coat

halves and cotyledons of Pisum sativum L., after uptake from a bathing

medium. The difference between sucrose and amino acids. J Plant

PhysioI120:419-429

196

Wolswinkel P (1985b) Phloem unloading and turgor-sensitive transport:

factors involved in sink control of assimilate partitioning. Physiol Plant

65:331-339

Wolswinkel P (1987a) The role of maternal seed tissues in the sink control of

asimilate transport into developing seeds. Plant Physiol Biochem

25:557-566

Wolswinkel P (1987b) Mechanisms of loading and unloading of xylem and

phloem. In OW Newman, KG Wilson, eds, Models in Plant Physiology

and Biochemistry, Vol 3. CRC Press, Boca Raton, pp 21-24

Wolswinkel P (1988) Nutrient transport into developing seeds. In lSI Atlas of

Science: Plants & Animals, Vol 1. Institute for Scientific Information,

Philadelphia, pp 298-302

Wolswinkel P, Ammerlaan A (1983) Phloem unloading in developing seeds of

Vicia faba L. the effect of several inhibitors on the release of sucrose and

amino acids by the seed coat. Planta 158:205-215

Wolswinkel P, Ammerlaan A (1984) Turgor-sensitive sucrose and amino acid

transport into developing seeds of Pisum sativum. Effect of a high

sucrose or mannitol concentration in experiments with empty ovules.

Physiol Plant 61 :172-182

Wolswinkel P, Ammerlaan A (1985a) Characteristics of sugar, amino acid,

and phosphate release from the seed coat and developing seeds of

Vicia faba and Pisum sativum. J Exp Bot 36:359-368

Wolswinkel P, Ammerlaan A (1985b) Effect of potassium on sucrose and

amino acid release from the seed coat of developing seeds of Pisum

sativum. Ann Bot 56:35-43

Wolswinkel P, Ammerlaan A (1986) Turgor-sensitive transport in developing

seeds of legumes: the role of the stage of development and the use of

excised vs. attached seed coats. Plant, Cell Environ 9: 133-140

197

Wolswinkel P, Ammerlaan A (1988) Effect of the osmotic environment of the seed coat on sucrose and amino acid transport into developing seeds of Lunaria annua, Acer pseudoplatanus and Glycine max. Physiol Plant 74:262-269

Wolswinkel P, Ammerlaan A (1989) Effect of the osmotic environment on assimilate transport in isolated developing embryos of maize (Zea mays). Ann Bot 63:705-708

Wolswinkel P, Ammerlaan A, Kuyvenhoven H (1983) Effect of KCN and p­chloromercuribenzenesulfonic acid on the release of sucrose and 2-

amino(1-14C)isobutyric acid by the seed coat of Pisum sativum. Physiol Plant 59:375-386

Wolswinkel P, Kraus E, Ammerlaan A (1986) Effect of the osmotic

environment on the balance between uptake and release of sucrose and amino acids by the seed coat and cotyledons of developing seeds of Pisum sativum. 37:1462-1471

Wright KM, Oparka KJ (1989) Sucrose uptake and partitioning in discs

derived from source versus sink potato tubers. Planta 177:237-244

Wyse RE (1986) Sinks as determinants of assimilate partitioning: possible sites

for regulation. In J Cronshaw, WJ Lucas, RJ Giaquinta, eds, Phloem Transport, Plant Biology, Vol 1. AR Liss, New York, pp 197-209

Wyse RE, zamski E, Tomos AD (1986) Turgor regulation of sucrose transport

in sugar beet taproot. Plant Physiol 81 :478-81

Yaklich RW, Wergin WP, Virgil EL (1986) Special secretory cells of the soybean seed coat. Protoplasma 134:78-87

198

Yamagata M, Kouchl H, Yoneyama T (1987) Partitioning and utilization of

photosynthate produced at different growth stages after anthesis in

soybean (Glycine max L. Merr.): analysis by long-term 13C-labelling

experiments. J Exp Bot 38: 1247-1259

Yeung EC (1983) Developmental changes in the branched parenchyma cells

of bean-seed coat. Protoplasma 118:225-229

Yeung EC, Brown DCW (1982) The osmotic environment of developing

embryos of Phaseolus vulgaris. Z Pflanzenphysiol 106: 149-156

Zierler K (1981) A critique of compartmental analysis. Ann Rev Biophys Bioeng

10:531-562